239 results on '"vectorisation"'
Search Results
2. A DSL and MLIR Dialect for Streaming and Vectorisation
- Author
-
da Silva, Manuel Cerqueira, Sousa, Luís, Paulino, Nuno, Bispo, João, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Skliarova, Iouliia, editor, Brox Jiménez, Piedad, editor, Véstias, Mário, editor, and Diniz, Pedro C., editor
- Published
- 2024
- Full Text
- View/download PDF
3. A Graph-Based Relook Beyond Metadata for Music Recommendation
- Author
-
Bharadwaj, Vishal, Mysore, Aravind S., Sangli, Ninad, Bharadwaj, Shraddha, Das, Bhaskarjyoti, Kacprzyk, Janusz, Series Editor, Gomide, Fernando, Advisory Editor, Kaynak, Okyay, Advisory Editor, Liu, Derong, Advisory Editor, Pedrycz, Witold, Advisory Editor, Polycarpou, Marios M., Advisory Editor, Rudas, Imre J., Advisory Editor, Wang, Jun, Advisory Editor, Roy, Satyabrata, editor, Sinwar, Deepak, editor, Dey, Nilanjan, editor, Perumal, Thinagaran, editor, and Tavares, João Manuel R. S., editor
- Published
- 2023
- Full Text
- View/download PDF
4. Multi-Stage Edge Detection for Generative Spatial Robotic Artwork
- Author
-
Nag, Sukanya, Bhattacharjee, Deepsikha, Bhaumik, Archisman, Deb, Suman, Angrisani, Leopoldo, Series Editor, Arteaga, Marco, Series Editor, Panigrahi, Bijaya Ketan, Series Editor, Chakraborty, Samarjit, Series Editor, Chen, Jiming, Series Editor, Chen, Shanben, Series Editor, Chen, Tan Kay, Series Editor, Dillmann, Rüdiger, Series Editor, Duan, Haibin, Series Editor, Ferrari, Gianluigi, Series Editor, Ferre, Manuel, Series Editor, Hirche, Sandra, Series Editor, Jabbari, Faryar, Series Editor, Jia, Limin, Series Editor, Kacprzyk, Janusz, Series Editor, Khamis, Alaa, Series Editor, Kroeger, Torsten, Series Editor, Li, Yong, Series Editor, Liang, Qilian, Series Editor, Martín, Ferran, Series Editor, Ming, Tan Cher, Series Editor, Minker, Wolfgang, Series Editor, Misra, Pradeep, Series Editor, Möller, Sebastian, Series Editor, Mukhopadhyay, Subhas, Series Editor, Ning, Cun-Zheng, Series Editor, Nishida, Toyoaki, Series Editor, Oneto, Luca, Series Editor, Pascucci, Federica, Series Editor, Qin, Yong, Series Editor, Seng, Gan Woon, Series Editor, Speidel, Joachim, Series Editor, Veiga, Germano, Series Editor, Wu, Haitao, Series Editor, Zamboni, Walter, Series Editor, Zhang, Junjie James, Series Editor, Mekhilef, Saad, editor, Shaw, Rabindra Nath, editor, and Siano, Pierluigi, editor
- Published
- 2022
- Full Text
- View/download PDF
5. [Nanomedicines for the treatment of serious diseases and dual research].
- Author
-
Couvreur P
- Published
- 2024
- Full Text
- View/download PDF
6. Developing an Efficient Vector-Friendly Implementation of the Breadth-First Search Algorithm for NEC SX-Aurora TSUBASA
- Author
-
Afanasyev, Ilya V., Voevodin, Vladimir V., Komatsu, Kazuhiko, Kobayashi, Hiroaki, Filipe, Joaquim, Editorial Board Member, Ghosh, Ashish, Editorial Board Member, Prates, Raquel Oliveira, Editorial Board Member, Zhou, Lizhu, Editorial Board Member, Kotenko, Igor, Editorial Board Member, Sokolinsky, Leonid, editor, and Zymbler, Mikhail, editor
- Published
- 2020
- Full Text
- View/download PDF
7. Evaluating the Effectiveness of a Vector-Length-Agnostic Instruction Set
- Author
-
Poenaru, Andrei, McIntosh-Smith, Simon, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Malawski, Maciej, editor, and Rzadca, Krzysztof, editor
- Published
- 2020
- Full Text
- View/download PDF
8. Exploring vectorisation for parallel breadth-first search on an advanced vector processor
- Author
-
Paredes Lopez, Mireya and Riley, Graham
- Subjects
511 ,graph algorithms ,graph 500 benchmark ,breadth first search ,parallel architecture ,vectorisation ,Xeon Phi - Abstract
Modern applications generate a massive amount of data that is challenging to process or analyse. Graph algorithms have emerged as a solution for the analysis of this data because they can represent the entities participating in the generation of large scale datasets in terms of vertices and their relationships in terms of edges. Graph analysis algorithms are used for finding patterns within these relationships, aiming to extract information to be further analysed. The breadth-first search (BFS) is one of the main graph search algorithms used for graph analysis and its optimisation has been widely researched using different parallel computers. However, the BFS parallelisation has been shown to be chal- lenging because of its inherent characteristics, including irregular memory access patterns, data dependencies and workload imbalance, that limit its scalability. This thesis investigates the optimisation of the BFS on the Xeon Phi, which is a modern parallel architecture provided with an advanced vector processor using a self-created development framework integrated with the Graph 500 benchmark. As a result, optimised parallel versions of two high-level algorithms for BFS were created using vectorisation, starting with the conventional top-down BFS algorithm and, building on this, leading to the hybrid BFS algorithm. The best implementations resulted in speedups of 1.37x and 1.33x, for a one million vertices graph, compared to the state-of-the-art, respectively. The hybrid BFS algorithm can be further used by other graph analysis algorithms and the lessons learned from vectorisation can be applied to other algorithms targeting the existing and future models of the Xeon Phi and other advanced vector architectures.
- Published
- 2017
9. Optimisation of the Microencapsulation of an Active Ingredient by Crosslinking and the Coating Method to Target Colon Diseases
- Author
-
Mounir Hammoudi, Djamel Atsamnia, Khaled Otmanine, Riadh Moumen, and Mustapha Oumouna
- Subjects
microcapsules ,colon diseases ,alginate ,chitosan ,vectorisation ,coating ,crosslinking ,Chemistry ,QD1-999 - Abstract
The aim of this study was to prepare microcapsules based on a natural polymer chitosan solution (high degree of deacetylation (DDA), low molecular weight (MW), and low viscosity)/sodium alginate in the presence of a crosslinking agent (glutaraldehyde), in order to encapsulate and vectorise the active principle towards the diseased organ (colon), without being diffused into other levels of the digestive tract, to increase the therapeutic effectiveness of treatment by chemotherapy and to reduce undesirable effects. The method of preparation of the microcapsules obtained from the sodium alginate/chitosan solution/active ingredients system was examined by conventional optical microscopy. In addition, an in vitro study was carried out on the active ingredients’ release profiles, depending on the pH simulating the gastric and intestinal media for the seven systems proposed. It should be mentioned that, in the basic medium (pH(colon) = 8), the release of the active ingredients is of the utmost importance. Nevertheless, control of this release can be improved by a crosslinking agent and the coating method. The dry [sodium alginate / chitosan solution / active ingredients + crosslinking 2 %] formulation coated with non-crosslinked chitosan (Formulation 7) is the standard formula that meets all the criteria from our earlier work, with a core release rate of 67 %. The PSD was unimodal, with sizes ranging from 750 µm to 900 µm.
- Published
- 2021
- Full Text
- View/download PDF
10. Vector Operations for Accelerating Expensive Bayesian Computations - A Tutorial Guide.
- Author
-
Warne, David J., Sisson, Scott A., and Drovandi, Christopher
- Subjects
BAYESIAN analysis ,SIGNAL-to-noise ratio ,MONTE Carlo method ,ARITHMETIC ,EXPERIMENTAL design - Abstract
Many applications in Bayesian statistics are extremely computationally intensive. However, they are often inherently parallel, making them prime targets for modern massively parallel processors. Multi-core and distributed computing is widely applied in the Bayesian community, however, very little attention has been given to fine-grain parallelisation using single instruction multiple data (SIMD) operations that are available on most modern CPUs. In this work, we practically demonstrate, using standard programming libraries, the utility of the SIMD approach for several topical Bayesian applications. Using the C programming language, we show that SIMD can improve the single-core floating point arithmetic performance by up to a factor of 6x compared scalar C code and more than 25 x compared with optimised R code. Such improvements are multiplicative to any gains achieved through multi-core processing. We illustrate the potential of SIMD for accelerating Bayesian computations and provide the reader with techniques for exploiting modern massively parallel processing environments. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
11. Fast DEM Collision Checks on Multicore Nodes
- Author
-
Krestenitis, Konstantinos, Weinzierl, Tobias, Koziara, Tomasz, Hutchison, David, Series Editor, Kanade, Takeo, Series Editor, Kittler, Josef, Series Editor, Kleinberg, Jon M., Series Editor, Mattern, Friedemann, Series Editor, Mitchell, John C., Series Editor, Naor, Moni, Series Editor, Pandu Rangan, C., Series Editor, Steffen, Bernhard, Series Editor, Terzopoulos, Demetri, Series Editor, Tygar, Doug, Series Editor, Weikum, Gerhard, Series Editor, Wyrzykowski, Roman, editor, Dongarra, Jack, editor, Deelman, Ewa, editor, and Karczewski, Konrad, editor
- Published
- 2018
- Full Text
- View/download PDF
12. Optimisation of the Microencapsulation of an Active Ingredient by Crosslinking and the Coating Method to Target Colon Diseases.
- Author
-
Hammoudi, M., Atsamnia, D., Otmanine, K., Moumen, R., and Oumouna, M.
- Subjects
- *
COLON diseases , *MOLECULAR capsules , *SODIUM alginate , *BIOPOLYMERS , *POLYMER solutions , *MOLECULAR weights , *MICROENCAPSULATION - Abstract
The aim of this study was to prepare microcapsules based on a natural polymer chitosan solution (high degree of deacetylation (DDA), low molecular weight (MW), and low viscosity)/sodium alginate in the presence of a crosslinking agent (glutaraldehyde), in order to encapsulate and vectorise the active principle towards the diseased organ (colon), without being diffused into other levels of the digestive tract, to increase the therapeutic effectiveness of treatment by chemotherapy and to reduce undesirable effects. The method of preparation of the microcapsules obtained from the sodium alginate/chitosan solution/active ingredients system was examined by conventional optical microscopy. In addition, an in vitro study was carried out on the active ingredients’ release profiles, depending on the pH simulating the gastric and intestinal media for the seven systems proposed. It should be mentioned that, in the basic medium (pH(colon) = 8), the release of the active ingredients is of the utmost importance. Nevertheless, control of this release can be improved by a crosslinking agent and the coating method. The dry [sodium alginate / chitosan solution / active ingredients + crosslinking 2 %] formulation coated with non-crosslinked chitosan (Formulation 7) is the standard formula that meets all the criteria from our earlier work, with a core release rate of 67 %. The PSD was unimodal, with sizes ranging from 750 µm to 900 µm. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
13. Exploiting Parallelism and Vectorisation in Breadth-First Search for the Intel Xeon Phi.
- Author
-
Paredes, Mireya, Riley, Graham, and Lujan, Mikel
- Subjects
- *
ALGORITHMS , *GRAPH algorithms , *PARALLEL processing , *PARALLEL computers , *MODERN architecture , *SEARCH algorithms , *PARALLEL programming - Abstract
Modern applications generate massive amounts of data that is challenging to process or analyse. Graph algorithms have emerged as a solution for the analysis of such data because they can represent the entities participating in the generation of large-scale datasets in terms of vertices and their relationships in terms of edges. Graph analysis algorithms are used for finding patterns within these relationships, aiming to extract information to be further analysed. The breadth-first search (BFS) is one of the main graph search algorithms used for graph analysis and its optimisation has been widely researched using different parallel computers. However, the parallelisation of BFS has been shown to be challenging because of its inherent characteristics, including irregular memory access patterns, data dependencies and workload imbalance, that limit its scalability. This paper investigates the optimisation of the BFS on the Xeon Phi (Knights Corner), a modern parallel architecture provided with an advanced vector processor supporting the AVX-512 instruction set, using a bespoke development framework integrated with the Graph 500 benchmark. In addition, to demonstrate portability, we show results for a direct port of the algorithms to a more recent version of the Xeon Phi (Knights Landing) and to a Skylake CPU which supports most of the AVX-512 instruction set. Optimised parallel versions of two high-level algorithms for BFS were created using vectorisation, starting with the conventional top-down BFS algorithm and, building on this, a hybrid BFS algorithm. On the KNC our best implementations result in speedups of 1.37x (top-down) and 1.37x (hybrid), for a one million vertices graph, compared to the state-of-the-art. On the KNL and Skylake, the performance is higher than on KNC. In addition, we show results of our best hybrid algorithm on real-world graphs from the SNAP datasets with speedups up to 1.3x on KNC. Performance on KNL and Skylake is again higher, demonstrating the robustness and portability of our algorithm. The hybrid BFS algorithm can be further used to speed up other graph analysis algorithms and the lessons learned from vectorisation can be applied to other algorithms targeting existing and future models of the Xeon Phi and other advanced vector architectures. [ABSTRACT FROM AUTHOR]
- Published
- 2020
- Full Text
- View/download PDF
14. A Study on Vectorisation and Paralellisation of the Monotonicity Approach
- Author
-
Skalna, Iwona, Duda, Jerzy, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Wyrzykowski, Roman, editor, Deelman, Ewa, editor, Dongarra, Jack, editor, Karczewski, Konrad, editor, Kitowski, Jacek, editor, and Wiatr, Kazimierz, editor
- Published
- 2016
- Full Text
- View/download PDF
15. An Algebraic Approach to Vectorial Programs
- Author
-
Charles Paperman and Sylvain Salvati and Claire Soyez-Martin, Paperman, Charles, Salvati, Sylvain, Soyez-Martin, Claire, Charles Paperman and Sylvain Salvati and Claire Soyez-Martin, Paperman, Charles, Salvati, Sylvain, and Soyez-Martin, Claire
- Abstract
Vectorial programming, the combination of SIMD instructions with usual processor instructions, is known to speed-up many standard algorithms. Simple regular languages have benefited from this technology. This paper is a first step towards pushing these benefits further. We take advantage of the inner algebraic structure of regular languages and produce high level representations of efficient vectorial programs that recognize certain classes of regular languages. As a technical ingredient, we establish equivalences between classes of vectorial circuits and logical formalisms, namely unary temporal logic and first order logic. The main result is the construction of compilation procedures that turns syntactic semigroups into vectorial circuits. The circuits we obtain are small in that they improve known upper-bounds on representations of automata within the logical formalisms. The gain is mostly due to a careful sharing of sub-formulas based on algebraic tools.
- Published
- 2023
- Full Text
- View/download PDF
16. An Algebraic Approach to Vectorial Programs
- Author
-
Paperman, Charles, Salvati, Sylvain, Soyez-Martin, Claire, Linking Dynamic Data (LINKS), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), and Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)
- Subjects
Vectorisation ,Theory of computation → Formal languages and automata theory ,[INFO]Computer Science [cs] ,Semigroups ,Automata theory - Abstract
Vectorial programming, the combination of SIMD instructions with usual processor instructions, is known to speed-up many standard algorithms. Simple regular languages have benefited from this technology. This paper is a first step towards pushing these benefits further. We take advantage of the inner algebraic structure of regular languages and produce high level representations of efficient vectorial programs that recognize certain classes of regular languages. As a technical ingredient, we establish equivalences between classes of vectorial circuits and logical formalisms, namely unary temporal logic and first order logic. The main result is the construction of compilation procedures that turns syntactic semigroups into vectorial circuits. The circuits we obtain are small in that they improve known upper-bounds on representations of automata within the logical formalisms. The gain is mostly due to a careful sharing of sub-formulas based on algebraic tools., LIPIcs, Vol. 254, 40th International Symposium on Theoretical Aspects of Computer Science (STACS 2023), pages 51:1-51:23
- Published
- 2023
- Full Text
- View/download PDF
17. Vectorisation of male supremac(h)ist ideologies in high-impact narratives and sociopolitical statism in western contemporary patriarchies
- Author
-
Yagüe-Pasamón, Sergio and Rivas Carmona, Mª del Mar
- Subjects
Idelologies ,Patriarchies ,Vectorisation ,Male Supremachist ,Sociopolitical Statism - Abstract
The present thesis dissertation, oriented at the fulfilment of the requirements for the obtention of the Philosophy Doctor, aims provide an intensive analysis on the deployment of traditional and contemporary narrative formats, in the context of the patriarchal establishments which have governed the Western world since early foundational civilisations, for the endurance of their supremac(h)ist organisation principles through the creation and maintenance of a conceptually recreated gender breach. The appropriate obtention of relevant data, for the surveyal of the potential encoding of gender unequal ideological concerns, as oriented to the alterity portrayal of the male and the female with socio-political stratification purposes, from the analysis of such a wide variety of narrative formats entails the implementation of diverse, text-specific, methodological guidelines. The technology in the contemporary world has imposed the inevitable presence of penetrating audiovisual ideological narratives with a massive outreach. Such would be the case of television, whose brief, but precise, advertisement narrations are studied to intergenerationally perpetuate the patriarchal task distribution that displaces women to the private sphere for the performance of kitchen and cleansing duties in a world were women cannot be denied to work and efforts for conciliation are necessary, as exhibited in the analysed costumbrist spots by major and highly consumer-endorsed brands ‘Natillas La Lechera’ and ‘La Cocinera’, or ‘Lejía Conejo’ bleach, ‘Vileda Windomatic’ squeege and ‘Philips Perfect Care’. As opposed to television advertising, other audiovisual formats, such as animation, do not intend to depict spectators’ daily life, but to present easy-to-embrace characters, who are studied to hold a powerful faculty to channelise narrative-controlled feelings and ideas. Such a factor would justify the eligibility of The Simpsons, as the longest scripted show in TV history and the most popular family in UK, over the royal family. In ‘The Diatribe of a Mad Housewife’, the female incorporation to the public sphere of professional life, which materialises in female writing, is deceitfully conceived as a source of misandry and conflict between the male figures of reference. Another audiovisual narrative format for male chauvinist ideologisation, video games, would interestingly resort to the immersion of fictional worlds governed by representational rules tacitly accepted by players when interacting with such influential major entertainment products, so the player-avatar identification facilitates the extrapolation of the gender-based in-game representations to the players’ external reality. Specifically, top-sales Resident Evil 4 (2005) and Bioshock Infinite (2013) present their respective female protagonists, Ashley Graham and Elisabeth Comstock, as passive entities to possess and custody in the private sphere, being both the characters the unfortunate threat to the socio-political stability of the status quo when the private custody decays and their control is usurped by male figures who move them to other public or private spaces. As observed in Resident Evil 4, the reversibility of such a systemic unbalance may be feasible when the damsel in danger is rescued from the private captivity setting and the initial custody by the male president of the United States and head of the free patriarchal world is restored. However, the male custody of Elisabeth Comstock in Bioshock Infinite’s public sphere, which implies her presence outside the house prophylactic confinement, is contrastively portrayed as the catalyser of the unleash of female uncontrolled powers and the inevitable destruction of the malecreated civilisation. The first five games in Pokémon saga offer a peaceful alternative to Resident Evil 4 and Bioshock Infinite, where gender alterity is imposed by the necessary-to-progress interaction of the player with male and female flat stereotypical characters, and the peaceful coexistence of humans and monsters is guaranteed by a hierarchised power apparatus with a minimal female representation. On the contrary, contemporary written formats, along with the atemporal urban legend and news, are also examined to complement the holistic contribution performed by narratives for the creation of a multiplatform supremac(h)ist ideological apparatus. Thus, wide-outreach social networking interactions may be contemplated to resort to hoaxes, pseudo-informative narratives, for the fostering of gender-biased conceptions which justify the organisation of the community around the male figure. Specifically, Twitter is analysed to have been utilised by Spanish (ultra)right political forces to summon the population’s fear to massive concentrations in early COVID-19 crisis to depict Spanish 8M feminist activism and overtly feminist first-order female politicians as peril for the wellbeing of the community, which prioritises their ambitions to individuals’ health and integrity. In order to filter and retrieve the tweets to explore, term (8M, 8-M), chronological (March 6th – June 21st, 2020) and geolocation (Spain) Boolean operators have been implemented in Twitter internal browser. Boolean operators may also prove relevant for the discovery of news as a written narrative genre actively contaminated by writers’ ideological communicative intentions during text elaboration and design, as observed in the ideological justification and ‘laundering’ of the toxic model of possessive masculinity exhibited upon Will Smith’s aggressive restitution of the woman-channelised family name staining at 2022 Academy Awards. To these regards, the filtered journalistic reports on Smith’s assault on Chris Rock upon Jada Pinkett’s alopecia joke, in Spanish, English and Portuguese, highlight Pinkett’s passivity in favour of the male participants’ interventions, who are to impose offence on the defenceless woman, in the case of Rock, and deliver medieval-like justice, as providentially assigned by the Catholic divinity, for the effective exercise of the custody of the damsel in danger, as incarnated by Smith. In this context, contemporary broadcasting platforms reveal as powerful vectors of influencer-based toxic masculinities. The last of the written genres examined befits those high impact viral accounts, with adaptative contextual flexibility to enhance its relevance for the narratee, a sense of veracity and a cautionary ending which exhort narratees to vectorise the subliminally ideological precepts encoded in the story ‘DNA’. These atemporal narratives, extracted from Spanish and English language Ortí and Sampere (2017), Camacho (2005, 2007), Brown & Flynn (2003) and Brunvald’s (1999) anthologies are filtered to obtain urban legends which depict women ‘in action’ under no male custody, as in Camacho’s (2007) “Secado Rápido”, a story at the crossroads of gender and Elder age inferiority in the private sphere, where an unsupervised widow kills an innocent puppy while arrogantly trying to microwave-dry it after an unplanned bath; Brown and Flynn’s (1999) “Chain reaction”, where an uncivil, yet irrelevant, woman’s behaviour in a public transport space causes a massive disaster which redounds in the destruction of the male-created urban skyline; or Hayes’ (2013), where the female ambition’s to trespass women’s patriarchal limitations lead her to reach the top of the Catholic Church under the disguise of John Anglicus, in early 13th century, but meets the capital penalty when misguided by her lust and her subsequently unexpected public labor during a religious service in the surroundings of Rome’s colosseum and St. Clement’s church. La presente tesis doctoral, que está orientada a la satisfacción de los requisitos para la obtención del título de doctor en la rama de lenguas y culturas, pretende llevar a cabo un análisis intensivo de la utilización de formatos narrativos tradicionales y contemporáneos, en el contexto de los statu quo patriarcales que han gobernado el mundo occidental desde los albores de sus civilizaciones fundacionales, con el objeto de perdurar sus principios de organización ‘supremac(h)istas’ por medio de la creación y mantenimiento de una brecha de género de origen conceptual. La obtención adecuada de información relevante sobre la posible existencia de codificaciones ideológicas en base a la desigualdad de género, que estarían dirigidos a una descripción altérica de la mujer y el hombre para la estratificación sociopolítica, en tan amplia variedad de formatos narrativos, requiere la implementación de líneas metodológicas diversas y adaptadas al formato textual correspondiente. Las prestaciones tecnológicas del mundo contemporáneo han impuesto la presencia ineludible de narrativas ideológicas audiovisuales de largo alcance. Este sería el caso de la televisión y sus cortas narrativas publicitarias, que, aunque cortas pero precisas, contribuyen a la perpetuación intergeneracional de la distribución de tareas patriarcal, que desplazaría a la mujer a la esfera privada para la satisfacción de obligaciones, en lo que a la cocina y a la higienización del hogar se refiere, en un mundo en el que no se puede alienar el derecho a la mujer a trabajar y que impone un esfuerzo extra para la conciliación familiar, como se observa en los anuncios, de corte costumbrista, de productos de marcas punteras como ‘Natillas La Lechera’, ‘La Cocinera, ‘Lejía Conejo’, ‘Vileda Windomatic’ o ‘Philips Perfect Care’. Al contrario que en el caso de la publicidad televisiva, otros formatos audiovisuales, como la animación, no persiguen retratar la vida diaria del espectador, sino ofrecer personajes con los que simpatizar, dada su probada valía para la canalización de ideas y emociones en las narrativas en las que se enmarcan. Este factor justifica la elección de Los Simpsons, que se erigen como el espectáculo guionizado de mayor duración en la historia de la televisión, así como la familia más popular en Reino Unido, incluso por encima de la familia real del país. En ‘Diatriba de una ama de casa loca’, la incorporación de la mujer a la vida profesional de la esfera pública, que en el caso de Marge se materializa en la escritura femenina, se concibe, no exenta de engaño, como una fuente de misandria y conflicto entre las figuras masculinas de referencia en el entorno de social de Marge. Otro formato narrativo audiovisual de ideologización supremac(h)ista tendría su seno en los videojuegos, que ofrecen a los jugadores una experiencia inmersiva en la que, como fruto de la interacción con estos productos de entretenimiento de gran influencia, los jugadores aceptan tácitamente las reglas representacionales y operativas que gobiernan el mundo ficticio del juego. En particular, los títulos superventas Resident Evil 4 (2005) y Bioshock Infinite (2013), muestran a sus protagonistas femeninos respectivos, Ashley Graham and Elisabeth Comstock, como sujetos pasivos a poseer y custodiar en la esfera privada, suponiendo ambos personajes una amenaza para la estabilidad del statu quo cuando la custodia privada inicial decae y su control es apropiado indebidamente por otras figuras masculinas, que ejercerán una custodia forzosa de la mujer cautiva en otros contextos públicos y privados. Como se puede observar en Resident Evil 4¸el desequilibrio sistémico causado por dicho ‘traspaso’ solo reviste reversibilidad cuando se rescata a la ‘damisela en apuros’ del escenario privado donde se mantiene cautiva y se restaura la custodia inicial, a cargo del presidente de los Estados Unidos y líder del mundo patriarcal libre. Por el contrario, la custodia masculina de Elisabeth Comstock en la esfera pública de Bioshock Infinite, que implica inherentemente la presencia de la mujer fuera de su confinamiento profiláctico entre las cuatro paredes del hogar, se conceptualiza como el catalizador de la liberación de los poderes descontrolados de la misma y la destrucción inevitable de la civilización creada por la mano del hombre. Por otro lado, los cinco primeros títulos de la saga Pokémon ofrecen una alternativa pacífica al mundo ficticio de Resident Evil 4 y Bioshock Infinite, donde la interacción del jugador con personajes masculinos y femeninos estereotípicos y planos se impone como una necesidad para el progreso de los jugadores en la narrativa, y la convivencia pacífica entre seres humanos y monstruos viene de la mano de un aparato de poder jerarquizado en el que la presencia de la mujer es mínima. Por otro lado, los formatos escritos contemporáneos, junto con otros formatos atemporales como la noticia o la leyenda urbana, también son objeto de análisis, de cara al descubrimiento de la contribución holística de narrativas diversas a la creación de un aparato ideológico supremac(h)ista multiplataforma. En este contexto, se examina la manera en la que las interacciones en redes sociales de amplio alcance recurren a los bulos, narrativas pseudo-informativas, para la promoción de concepciones basadas en la desigualdad de género, que vendrían a justificar la organización sociopolítica de la comunidad en torno a la figura del hombre. En particular, se analiza el uso de Twitter, por parte de fuerzas de la (ultra)derecha política española, para avivar el miedo de la población general a las concentraciones masificadas en los albores de la crisis sanitaria de la COVID-19 y retratar, así, al activismo feminista español en torno al 8M, y a las políticas de primer orden que lo encabezan, como un peligro para el bienestar de la comunidad que priorizaría sus ambiciones a la salud e integridad de los ciudadanos. Para la recuperación y subsiguiente filtrado de los tuits a estudiar, se implementan los operadores booleanos terminológicos (8M, 8-M), cronológicos (del 6 de marzo al 21 de junio de 2020) y de geolocalización (España) en el buscador interno de Twitter. La pertinencia de los operadores Booleanos para el estudio de la noticia, como género narrativo que adolece de una contaminación por parte de las intenciones comunicativas ideológicas de sus escritores, durante el proceso de elaboración textual, se puede observar en la justificación ideológica y el ‘blanqueamiento’ del modelo de masculinidad tóxica exhibida por Will Smith en la restitución agresiva de la honra familiar, previamente violada por medio de la explotación de la mujer del núcleo familiar como instrumento para ello. En relación a esto, los relatos periodísticos filtrados en español, inglés y portugués destacan la pasividad de Pinkett en favor de la intervención de los intervinientes masculinos, quienes imponen su ofensa verbal sobre la mujer indefensa, en el caso de Rock, e imparten una justicia cuasi-medieval, de acuerdo a los designios providenciales de la divinidad católica, para el ejercicio de la custodia de la ‘damisela en apuros’, en el caso de Smith. En este contexto, las plataformas de retransmisión de contenidos contemporáneas se erigen como vectores poderosos de las masculinidades tóxicas exhibidas por las figuras influyentes antes mencionadas. El último de los géneros escritos explorados en la presente tesis correspondería a aquellos relatos virales de alto impacto, que destacarían por su flexibilidad contextual adaptativa al público meta, la sensación de veracidad que recrean y su final precautorio, factores que contribuirían, de forma holística, a la vectorización de los preceptos ideológicos imbricados subliminalmente en el ‘ADN’ de la narración. Estas narrativas atemporales, que han sido exploradas en las antologías en lengua castellana e inglesa de Ortí y Sampere (2017), Camacho (2005, 2007), Brown y Flynn (2003) y Brunvald (1999), han sido sometidas, con posterioridad, a su filtrado para la obtención de leyendas urbanas que crean un retrato distorsionado de la mujer ‘en acción’, lejos de la custodia masculina, como sería el caso de “Secado Rápido” (Camacho, 2007), una historia que entremezcla género y edadismo, como factores clave en la inferioridad del individuo, en la que una viuda sin supervisión acaba con la vida de un cachorro de perro al intentar, en un alarde de inteligencia arrogante, secarlo en el microondas tras un baño que no estaba en los planes de la ‘abuelita’. Así mismo, en la leyenda urbana de título “Chain Reaction” (Brown and Flynn, 1999), el comportamiento tan inadecuado como inocuo de una mujer poco cívica en el espacio público del transporte colectivo origina la devastación del paisaje urbano civilizado creado por el patriarcado. Por último, también es objeto de análisis la leyenda urbana de Hayes (2013), en la que la ambición femenina lleva a una mujer a traspasar las limitaciones previstas por el patriarcado para su género y alcanzar así, a principios del siglo XIII y bajo la identidad de John Anglicus, la cima de la jerarquía católica, siendo presa de su lujuria, que asombrará a los allí presentes con un inesperado parto durante una procesión religiosa en las cercanías del Coliseo Romano y la iglesia de San Clemente, y su posterior ejecución, como único sino de la mujer que invade los espacios reservados para el hombre.
- Published
- 2023
18. Minimum Memory Vectorisation of Wavelet Lifting
- Author
-
Barina, David, Zemcik, Pavel, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Nierstrasz, Oscar, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Sudan, Madhu, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Vardi, Moshe Y., Series editor, Weikum, Gerhard, Series editor, Blanc-Talon, Jacques, editor, Kasinski, Andrzej, editor, Philips, Wilfried, editor, Popescu, Dan, editor, and Scheunders, Paul, editor
- Published
- 2013
- Full Text
- View/download PDF
19. Ferrocene Functionalized Endocrine Modulators as Anticancer Agents
- Author
-
Hillard, Elizabeth A., Vessières, Anne, Jaouen, Gerard, Jaouen, Gérard, editor, and Metzler-Nolte, Nils, editor
- Published
- 2010
- Full Text
- View/download PDF
20. Optimisation of the two-dimensional hydraulic model LISFOOD-FP for CPU architecture.
- Author
-
Neal, Jeffrey, Dunne, Toby, Sampson, Chris, Smith, Andrew, and Bates, Paul
- Subjects
- *
COASTAL zone management , *HYDRAULIC models , *FLOODS , *CENTRAL processing units , *MATHEMATICAL optimization , *MANAGEMENT - Abstract
Flood inundation models are increasingly used for a wide variety of river and coastal management applications. Nevertheless, the computational effort to run these models remains a substantial constraint on their application. In this study four developments to the LISFLOOD-FP 2D flood inundation model have been documented that: 1) refine the parallelisation of the model; 2) reduce the computational burden of dry cells; 3) reduce the data movements between CPU and RAM; and 4) vectorise the core numerical solver. The value of each of these developments in terms of compute time and parallel efficiency was tested on 12 test cases. For realistic test cases, improvements in single core performance of between 4.2x and 8.4x were achieved, which when combined with parallelisation on 16 cores resulted in computation times 34-60x shorter than previous LISFLOOD-FP models on one core. Results were compared to a sample of commercial models for context. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF
21. Synthèse et évaluation pharmacologique d'une nouvelle famille de transporteurs riches en guanidiniums
- Author
-
Lavoie, Christine, Marsault, Éric, Marouseau, Etienne, Boudreault, Pierre-Luc, Lavoie, Christine, Marsault, Éric, Marouseau, Etienne, and Boudreault, Pierre-Luc
- Abstract
Vectoriser les médicaments, c’est-à-dire les associer à un vecteur pour les amener là où ils doivent agir, présente un intérêt majeur. En effet, si la dose administrée va sélectivement à la cible, elle peut être diminuée : cela présente un intérêt économique et réduit les chances d’effets secondaires des suites d’une action ailleurs que là où elle est souhaitée. Pour les cibles intracellulaires, les peptides perméants sont prometteurs en raison de leur production aisée, de leur faible toxicité et de leur efficacité pour entrer dans les cellules. Cependant, leur sélectivité a été peu étudiée et leur internalisation est encore mal comprise et maîtrisée, ce qui limite l’efficacité de leurs applications. Ce projet visait à générer des analogues de peptides perméants cationiques pour étudier leur internalisation dans la cellule. Un autre objectif était de voir si une contrainte stérique pouvait créer une discrimination entre types cellulaires et générer des pistes pour un trans- port sélectif. Enfin, le dernier objectif était de créer des composés ayant de bons profils pharmacologiques. La thèse est divisée en trois parties entrecoupées de transitions : A. A partir de designs semi-rationnels, une série de composés analogues de peptides perméants cationiques a été produite. Ces résultats, publiés dans un premier article, ont permis de valider les méthodes de synthèse, le profil de toxicité et le mécanisme primaire d’internalisation des composés, ainsi que d’identifier le composé le plus efficace de la série. B. Par la suite, l’approche synthétique a été optimisée et plusieurs analogues structu- raux de la première série ont été produits. D’autres lignées cellulaires ont aussi été testées, dans le but d’observer une sélectivité dans l’internalisation. Ces résultats ainsi qu’une étude de la stabilité plasmatique de ces composés, bien que préliminaires, sont présentés. C. Enfin, d’autres structures de molécules relativement différentes ont été conceptua- lisées. Les synthèse
- Published
- 2022
22. Symbol Recognition Combining Vectorial and Statistical Features
- Author
-
Locteau, Hervé, Adam, Sébastien, Trupin, Éric, Labiche, Jacques, Héroux, Pierre, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Nierstrasz, Oscar, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Sudan, Madhu, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Dough, Series editor, Vardi, Moshe Y., Series editor, Weikum, Gerhard, Series editor, Liu, Wenyin, editor, and Lladós, Josep, editor
- Published
- 2006
- Full Text
- View/download PDF
23. Optimisation of the microencapsulation of an active ingredient by crosslinking and the coating method to target colon diseases
- Author
-
Riadh Moumen, Khaled Otmanine, Mustapha Oumouna, Djamel Atsamnia, and Mounir Hammoudi
- Subjects
Active ingredient ,vectorisation ,Chemistry ,General Chemical Engineering ,technology, industry, and agriculture ,coating ,macromolecular substances ,General Chemistry ,engineering.material ,Combinatorial chemistry ,microcapsules ,Coating ,engineering ,alginate ,crosslinking ,chitosan ,colon diseases ,QD1-999 - Abstract
The aim of this study was to prepare microcapsules based on a natural polymer chitosan solution (high degree of deacetylation (DDA), low molecular weight (MW), and low viscosity)/sodium alginate in the presence of a crosslinking agent (glutaraldehyde), in order to encapsulate and vectorise the active principle towards the diseased organ (colon), without being diffused into other levels of the digestive tract, to increase the therapeutic effectiveness of treatment by chemotherapy and to reduce undesirable effects. The method of preparation of the microcapsules obtained from the sodium alginate/chitosan solution/active ingredients system was examined by conventional optical microscopy. In addition, an in vitro study was carried out on the active ingredients’ release profiles, depending on the pH simulating the gastric and intestinal media for the seven systems proposed. It should be mentioned that, in the basic medium (pH(colon) = 8), the release of the active ingredients is of the utmost importance. Nevertheless, control of this release can be improved by a crosslinking agent and the coating method. The dry [sodium alginate / chitosan solution / active ingredients + crosslinking 2 %] formulation coated with non-crosslinked chitosan (Formulation 7) is the standard formula that meets all the criteria from our earlier work, with a core release rate of 67 %. The PSD was unimodal, with sizes ranging from 750 µm to 900 µm.
- Published
- 2021
- Full Text
- View/download PDF
24. FOREST COVER CHANGE OR MISINTERPRETATION? ON DEPENDENT AND INDEPENDENT VECTORISATION APPROACHES.
- Author
-
Bednarczyk, Barbara, Kaim, Dominik, and Ostafin, Krzysztof
- Subjects
- *
LAND cover , *HISTORICAL maps , *UNCERTAINTY - Abstract
The paper compares the influence of dependent and independent vectorisation approaches on forest cover change analysis, with a hypothesis that the former reduces the number and area of sliver polygons. Independent vectorisation is based on separate creation of the vector layer for each period in the time series, while the dependent is based on modification of the successive vector layers. The comparison is based on three map sets-the second Austrian military survey (1861/1862), a Polish military map (1936) and a Polish topographic map (1979) and carried out in Szczawnica commune located in the Polish Carpathians. The results show that the overall differences between the two vectorisation approaches are low at the commune level, but the local differences, within the grids 500 × 500 m might be up to 30-40 %. Statistical analysis did not indicate any considered variable directly responsible for the differences, confirming that it is a randomly distributed phenomenon. The results show also that the dependent vectorisation cannot eliminate the existence of sliver polygons, but their number may be limited when compared to the independent approach. As the dependent vectorisation is much more time efficient, we conclude that it might be a better solution in the situation when manual vectorisation is the most appropriate method of land use data acquisition from historical maps. [ABSTRACT FROM AUTHOR]
- Published
- 2016
- Full Text
- View/download PDF
25. Type-driven data layouts for improved vectorisation.
- Author
-
Šinkarovs, Artjoms and Scholz, Sven‐Bodo
- Subjects
DATA structures ,ALGORITHMS ,COMPUTER programmers ,VECTOR analysis ,MATHEMATICAL transformations - Abstract
Vector instructions of modern CPUs are crucially important for the performance of compute-intensive algorithms. Auto-vectorisation often fails because of an unfortunate choice of data layout by the programmer. This paper proposes a data layout inference for auto-vectorisation that identifies layout transformations that convert single instruction, multiple data-unfavourable layouts of data structures into favourable ones. We present a type system for layout transformations, and we sketch an inference algorithm for it. Finally, we present some initial performance figures for the impact of the inferred layout transformations. They show that non-intuitive layouts that are inferred through our system can have a vast performance impact on compute intensive programs. Copyright © 2015 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]
- Published
- 2016
- Full Text
- View/download PDF
26. User-Guided Vectorisation of Pixel Art through Spring Simulation
- Author
-
Matušovič, Marko (author) and Matušovič, Marko (author)
- Abstract
This research builds upon a previous method of vectorisation of pixel art by pixel neighbour connecting and boundary energy minimisation. The downside of the previous method is the lack of user input throughout the process and possible divergence between the results and the artist’s vision. The proposed method uses the first part of previous method to connect neighbour pixels and continues with a spring simulation of the boundaries. A process that can be heavily user-guided with adjusting the stiffness of the springs. Results show the proposed method to achieve higher variance in the results, possibly resulting in output images closer to the artist’s vision. However, introduce lengthy process for the user that can be removed by providing higher level GUI., CSE3000 Research Project, Computer Science and Engineering
- Published
- 2021
27. Nanoformulation des corps gras. Nanocapsules, nanoémulsions
- Author
-
Simonnet Jean-Thierry, Richart Pascal, Aubrun Odile, and L’alloret Florence
- Subjects
nanoemulsions ,Ostwald ripening ,depletion ,nanocapsules ,nanoprecipitation ,vectorisation ,Oils, fats, and waxes ,TP670-699 - Abstract
Nanocapsules (submicron particles, of diameter 250 nm, with an oily core enclosed in a solid polymeric shell) were prepared using a controlled nanoprecipitation process with preformed polymers (for example poly-E caprolactone, polyethylene adipate). Nanocapsules formed by this method contain up to 7.5% oil (1% polymer). The following properties are observed using nanocapsules : improved oxygen stability of active ingredients (retinol, β carotene), significantly improved vectorisation (up to 3 times) in skin and a reservoir effect. Due to their solid polymer shell nanocapsules are stable when formulated and penetrate intact into the superficial layers of the epidermis. Nanoemulsions wich are very fine oil-in-water dispersions having droplet diameter smaller than 100 nanometers, were prepared with a high shear device. Two major sources of unstability were identified and extensively studied: Ostwald ripening and depletion induced floculation following the addition of thickening polymers. The control of these two mechanisms allowed the industrial production of a large variety of cosmetic products. The nanoemulsions are easily valued in skin care due to their good sensorial properties and their biophysical properties.
- Published
- 2004
- Full Text
- View/download PDF
28. Hybrid multicore/vectorisation technique applied to the elastic wave equation on a staggered grid.
- Author
-
Titarenko, Sofya and Hildyard, Mark
- Subjects
- *
WAVE equation , *MATHEMATICAL transformations , *DISCRETIZATION methods , *PARTIAL differential equations , *FINITE differences , *ARRAY processors , *ELASTIC waves - Abstract
In modern physics it has become common to find the solution of a problem by solving numerically a set of PDEs. Whether solving them on a finite difference grid or by a finite element approach, the main calculations are often applied to a stencil structure. In the last decade it has become usual to work with so called big data problems where calculations are very heavy and accelerators and modern architectures are widely used. Although CPU and GPU clusters are often used to solve such problems, parallelisation of any calculation ideally starts from a single processor optimisation. Unfortunately, it is impossible to vectorise a stencil structured loop with high level instructions. In this paper we suggest a new approach to rearranging the data structure which makes it possible to apply high level vectorisation instructions to a stencil loop and which results in significant acceleration. The suggested method allows further acceleration if shared memory APIs are used. We show the effectiveness of the method by applying it to an elastic wave propagation problem on a finite difference grid. We have chosen Intel architecture for the test problem and OpenMP (Open Multi-Processing) since they are extensively used in many applications. [ABSTRACT FROM AUTHOR]
- Published
- 2017
- Full Text
- View/download PDF
29. Nanoparticles for the oral administration of amphotericin B for leishmaniasis treatment
- Author
-
Lipa Castro, Antonio, Institut Galien Paris-Saclay (IGPS), Institut de Chimie du CNRS (INC)-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Université Paris-Saclay, Gillian Barratt, and STAR, ABES
- Subjects
Amphotéricine B ,Vectorisation ,Voie orale ,[SDV.SP]Life Sciences [q-bio]/Pharmaceutical sciences ,[SDV.SP] Life Sciences [q-bio]/Pharmaceutical sciences ,Oral administration ,Cochléates ,Amphotericin B ,Drug delivery ,Leishmaniose ,[SDV.MP.PAR]Life Sciences [q-bio]/Microbiology and Parasitology/Parasitology ,Leishmaniasis ,[SDV.MP.PAR] Life Sciences [q-bio]/Microbiology and Parasitology/Parasitology ,Cochleate - Abstract
Leishmaniasis is a wide-spread tropical diseasewith an increasing number of cases despite preventive and therapeutic intervention. The treatment of choice is the antifungal Amphotercin B (AmB) that shows food efficacy and low resistance, but has low oral bioavailability and a small therapeutic window. In this work, a drug delivery system for AmB, cochleates, was developed as an oral treatment. Cochleates are assemblies of phospholipid bilayers rolled into cylinders bridged by divalent cations. In order to provide an affordable product for developping countries, a phospholipid of naturalorigin was used. The composition was optimized in terms of AmB encapsulation and protection in gastro-intestinal conditions. The cochleates were characterized by various physico-chemical techniques such as optical and electron microscopy, small-angel X-ray diffraction and circular dichroism. After a study of their interactions with Caco2 cells, pharmacokinetics and biodistribution in rats and efficacy against Leishmania donovani in mice were determined. The results are promising for the development of new therapy for Leishmaniasis., La leishmaniose est une maladie très répandue dans le monde dont, malgré les mesures de prévention et les traitements existants, le nombre de cas reste en augmentation. L’amphotericine B (AmB), antifongique et antileishmanienne, est la molécule de préférence à cause de son bonne efficacité et absence de résistance. Cependant sa faible biodisponibilité orale et son étroit fenêtre thérapeutique limitent son utilisation. Dans ce travail, un système de délivrance de l’AmB, les cochléates, a été développé afin de proposer un traitement efficace par voie orale. Les cochléates sont des assemblages des bicouches de phospholipide enroulées et stabilisées par des cations divalentes. Afin de faciliter l’accès aux pays en voie de développement, un phospholipide d’origine naturelle a été utilisé. La composition a été optimisée pour le taux d’encapsulation de l’AmB et sa protection en milieux gastro-intestinaux. Les cochléates ont été caractérisés par diverses techniques physico-chimiques telles que la microscopie optique électronique, la diffraction des rayons-X aux petits angles et le dichroïsme circulaire. Après une étude d’interaction avec les cellules Caco2, le profil pharmacocinétique et biodistribution chez le rat ainsi que l’efficacité sur Leishmania donovani chez la souris ont été déterminés. Les résultats sont prometteurs pour le développement d’une nouvelles thérapie anti-leishmanienne.
- Published
- 2021
30. Fast Quadratic Programming for Mean-Variance Portfolio Optimisation
- Author
-
Kontosakos, Vasileios E.
- Published
- 2020
- Full Text
- View/download PDF
31. Outils de calculs pour le HPDA : approche Cache-Oblivious et SIMD
- Author
-
Peou, Kenny, STAR, ABES, Laboratoire Interdisciplinaire des Sciences du Numérique (LISN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Agenium Scale, Systèmes Parallèles (ParSys), Algorithmes, Apprentissage et Calcul (AAC), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Interdisciplinaire des Sciences du Numérique (LISN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Université Paris-Saclay, and Joël Falcou
- Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI] ,Vectorisation ,Calcul Haute Performance (CHP) ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,Précision numérique ,[INFO.INFO-AO] Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,High Performance Computing (HPC) ,Numerical Precision ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] - Abstract
This work presents three contributions to the fields of CPU vectorization and machine learning. The first contribution is an algorithm for computing an average with half precision floating point values. In this work performed with limited half precision hardware support, we use an existing software library to emulate half precision computation. This allows us to compare the numerical precision of our algorithm to various commonly used algorithms. Finally, we perform runtime performance benchmarks using single and double floating point values in order to anticipate the potential gains from applying CPU vectorization to half precision values. Overall, we find that our algorithm has slightly worse best-case numerical performance in exchange for significantly better worst-case numerical performance, all while providing similar runtime performance to other algorithms. The second contribution is a fixed-point computational library designed specifically for CPU vectorization. Existing libraries fail rely on compiler auto-vectorization, which fail to vectorize arithmetic multiplication and division operations. In addition, these two operations require cast operations which reduce vectorizability and have a real computational cost. To allevieate this, we present a fixed-point data storage format that does not require any cast operations to perform arithmetic operations. In addition, we present a number of benchmarks comparing our implementation to existing libraries and present the CPU vectorization speedup on a number of architectures. Overall, we find that our fixed point format allows runtime performance equal to or better than all compared libraries. The final contribution is a neural network inference engine designed to perform experiments varying the numerical datatypes used in the inference computation. This inference engine allows layer-specific control of which data types are used to perform inference. We use this level of control to perform experiments to determine how aggressively it is possible to reduce the numerical precision used in inferring the PVANet neural network. In the end, we determine that a combination of the standardized float16 and bfloat16 data types is sufficient for the entire inference., Ce travail présente trois contributions aux domaines de la vectorisation des CPU et de l'apprentissage automatique. La première contribution est un algorithme pour calculer une moyenne avec des valeurs en virgule flottante de demi-précision. Dans ce travail réalisé avec un support matériel de demi-précision limité, nous utilisons une bibliothèque logicielle existante pour émuler le calcul de demi-précision. Cela nous permet de comparer la précision numérique de notre algorithme à celle de divers algorithmes couramment utilisés. Enfin, nous effectuons des tests de performance d'exécution en utilisant des valeurs à virgule flottante simples et doubles afin d'anticiper les gains potentiels de l'application de la vectorisation du CPU aux valeurs de demi-précision. Dans l'ensemble, nous constatons que notre algorithme présente des performances numériques légèrement inférieures dans le meilleur des cas en échange de performances numériques nettement supérieures dans le pire des cas, tout en offrant des performances d'exécution similaires à celles d'autres algorithmes. La deuxième contribution est une bibliothèque de calcul en virgule fixe conçue spécifiquement pour la vectorisation du CPU. Les bibliothèques existantes ne reposent pas sur l'auto-vectorisation du compilateur, qui ne parvient pas à vectoriser les opérations arithmétiques de multiplication et de division. De plus, ces deux opérations nécessitent des opérations de cast qui réduisent la vectorisabilité et ont un réel coût de calcul. Pour remédier à ce problème, nous présentons un format de stockage de données en virgule fixe qui ne nécessite aucune opération de cast pour effectuer des opérations arithmétiques. De plus, nous présentons un certain nombre de benchmarks comparant notre implémentation aux bibliothèques existantes et nous présentons la vitesse de vectorisation du CPU sur un certain nombre d'architectures. Dans l'ensemble, nous constatons que notre format en virgule fixe permet des performances d'exécution égales ou supérieures à toutes les bibliothèques comparées. La dernière contribution est un moteur d'inférence de réseau neuronal conçu pour réaliser des expériences en variant les types de données numériques utilisées dans le calcul d'inférence. Ce moteur d'inférence permet un contrôle spécifique à la couche des types de données utilisés pour effectuer l'inférence. Nous utilisons ce niveau de contrôle pour réaliser des expériences visant à déterminer l'agressivité avec laquelle il est possible de réduire la précision numérique utilisée dans l'inférence du réseau neuronal PVANet. Au final, nous déterminons qu'une combinaison des types de données standardisés float16 et bfoat16 est suffisante pour l'ensemble de l'inférence.
- Published
- 2021
32. Utilisation d’un Système d’Information Géographique pour représenter la distribution spatiale de composés chimiques ou organiques selon le type de compartiments d’un organisme. Exemple de la répartition de la concentration de glucides non structuraux dans un arbre
- Author
-
Thirion, Erwin, SILVA (SILVA), and AgroParisTech-Université de Lorraine (UL)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)
- Subjects
programmation ,[SDV.EE]Life Sciences [q-bio]/Ecology, environment ,vectorisation ,illustration ,[SDE]Environmental Sciences ,script ,SIG - Abstract
International audience; La gestion de données spatiales et leur représentation géographique s’est énormément développée grâce à l’essor des Systèmes d’Informations Géographiques (SIG). Cet article propose d’utiliser les fonctionnalités de ces outils pour créer une illustration afin de représenter spatialement l’intensité d’un processus ou la distribution d’un composé au sein d’un organisme vivant (comme par exemple les parties aériennes et souterraines, organes d’âges différents …). Cette tâche n’est pas aisée car aucun outil spécifique n’est disponible. Cet article présente une méthode reposant sur le logiciel SIG ArcGIS et les librairies SIG de R pour réaliser cette représentation d’information à l’aide d’une couche d’information de type vecteur (découpage en entités distinctes qui s’oppose au type raster qui représente l’espace dans un maillage régulier). Techniquement, la démarche proposée consiste 1) à utiliser un SIG (ArcGIS) pour vectoriser une image représentant l’organisme, puis 2) produire une illustration à partir de la couche vecteur créée selon le découpage souhaité. Plus particulièrement, l’image choisie pour représenter l’organisme étudié est vectorisée en objets de type polygone correspondant à des compartiments d’intérêt de l’organisme. Enfin la représentation de l’intensité du phénomène selon les différents compartiments est accomplie à l’aide de deux logiciels, ArcGIS et les librairies SIG du logiciel libre de statistique R. Ce deuxième présente l’avantage, une fois le fichier de forme créé, de pouvoir être implémenté dans un script d’analyse de données R, logiciel utilisé couramment dans la communauté scientifique. Cette seconde approche a permis, dans l’exemple présenté dans cet article, de représenter la répartition des concentrations en azote ou en réserves glucidiques dans les différents compartiments d’un hêtre à différentes dates et soumis à des traitements contrastés
- Published
- 2021
33. Conception, synthèse et évaluation d’agents perturbateurs vasculaires pour le traitement de l’arthrose et des douleurs arthrosiques
- Author
-
Prudhomme, Vassili, STAR, ABES, Institut de Chimie de Clermont-Ferrand (ICCF), Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Université Clermont Auvergne (UCA)-Institut national polytechnique Clermont Auvergne (INP Clermont Auvergne), Université Clermont Auvergne (UCA)-Université Clermont Auvergne (UCA), Université Clermont Auvergne, and Sylvie Ducki
- Subjects
Indoles ,Angiogenèse ostéochondrale ,Agents perturbateurs vasculaires ,Pain ,Hybrids ,Douleur ,Vaisseaux sanguins ,Blood vessels ,Articulations ,Tubulin ,Osteoarthritis ,Vectorization ,Arthrose ,[SDV.EE.SANT] Life Sciences [q-bio]/Ecology, environment/Health ,Hybrides ,Tubuline ,Hybridization ,[SDV.EE.SANT]Life Sciences [q-bio]/Ecology, environment/Health ,Inflammation ,Vascular disrupting agents ,Vectorisation ,[CHIM.ORGA]Chemical Sciences/Organic chemistry ,Hybridation ,Osteochondral angiogenesis ,[CHIM.ORGA] Chemical Sciences/Organic chemistry ,Cartilage ,Joints - Abstract
Osteoarthritis is a chronic disease leading to the progressive deterioration of joint cartilage. This irreversible degradation causes pain and mobility issues. More than 60% of French people over 65 are affected by this disease, which represents 10% of the global population. Current treatments focus on reducing pain, but none can stop the progression of arthritis. An innovative strategy proposes to target osteochondral angiogenesis. Pathological vascularization process in the joints plays an important role in cartilage deterioration. We designed and synthesized several molecular hybrids by combining the pharmacophores of various known vascular disrupting agents such as combretastatin A-4, phenstatin or E7010. The antivascular activity of these « hybrids » was finally evaluated in vitro. Some have shown promising results., L’arthrose est une maladie chronique des articulations se manifestant par l’altération progressive du cartilage articulaire. Cette dégradation irréversible engendre chez le patient des douleurs aiguës voire chroniques, ainsi qu’une difficulté à se déplacer. Plus de 60% des Français de plus de 65 ans sont concernés par cette maladie, soit 10% de la population. Les traitements actuels consistent essentiellement à réduire la douleur du patient, mais aucun ne permet d’inhiber ou d’enrayer la progression de la maladie. Une stratégie innovante, actuellement en développement, propose de cibler l’angiogenèse ostéochondrale. En effet, il a été démontré que ce processus de vascularisation, indésirable au niveau des articulations, jouait un rôle important dans la dégradation du cartilage. Nous avons conçu et synthétisé différents hybrides moléculaires, dont les structures combinent les pharmacophores de divers agents perturbateurs vasculaires issus de la littérature telles que la combretastatine A-4, la phenstatine, ou encore E7010. L’activité antivasculaire de ces molécules « hybrides » a finalement été évaluée in vitro. Certaines présentent des résultats prometteurs.
- Published
- 2021
34. Tuning a Finite Difference Computation for Parallel Vector Processors.
- Author
-
Zumbusch, Gerhard
- Abstract
Current CPU and GPU architectures heavily use data and instruction parallelism at different levels. Floating point operations are organised in vector instructions of increasing vector length. For reasons of performance it is mandatory to use the vector instructions efficiently. Several ways of tuning a model problem finite difference stencil computation are discussed. The combination of vectorisation and an interleaved data layout, cache aware algorithms, loop unrolling, parallelisation and parameter tuning lead to optimised implementations at a level of 90% peak performance of the floating point pipelines on recent Intel Sandy Bridge and AMD Bulldozer CPU cores, both with AVX vector instructions as well as on Nvidia Fermi/ Kepler GPU architectures. Furthermore, we present numbers for parallel multi-core/ multi-processor and multi-GPU configurations. They represent regularly more than an order of speed up compared to a standard implementation. The analysis may also explain deficiencies of automatic vectorisation for linear data layout and serve as a foundation of efficient implementations of more complex expressions. [ABSTRACT FROM PUBLISHER]
- Published
- 2012
- Full Text
- View/download PDF
35. Block Fusion on Dynamically Adaptive Spacetree Grids for Shallow Water Waves.
- Author
-
Weinzierl, Tobias, Bader, Michael, Unterweger, Kristof, and Wittmann, Roland
- Subjects
- *
CARTESIAN coordinates , *SIMULATION methods & models , *NUMERICAL analysis , *COORDINATES , *MATHEMATICAL analysis - Abstract
Spacetrees are a popular formalism to describe dynamically adaptive Cartesian grids. Even though they directly yield a mesh, it is often computationally reasonable to embed regular Cartesian blocks into their leaves. This promotes stencils working on homogeneous data chunks. The choice of a proper block size is sensitive. While large block sizes foster loop parallelism and vectorisation, they restrict the adaptivity's granularity and hence increase the memory footprint and lower the numerical accuracy per byte. In the present paper, we therefore use a multiscale spacetree-block coupling admitting blocks on all spacetree nodes. We propose to find sets of blocks on the finest scale throughout the simulation and to replace them by fused big blocks. Such a replacement strategy can pick up hardware characteristics, i.e. which block size yields the highest throughput, while the dynamic adaptivity of the fine grid mesh is not constrained-applications can work with fine granular blocks. We study the fusion with a state-of-the-art shallow water solver at hands of an Intel Sandy Bridge and a Xeon Phi processor where we anticipate their reaction to selected block optimisation and vectorisation. [ABSTRACT FROM AUTHOR]
- Published
- 2014
- Full Text
- View/download PDF
36. Synthesis and study of compounds able to activate MAIT cells
- Author
-
Yvorra, Thomas, Chimie biologique des membranes et ciblage thérapeutique (CBMCT - UMR 3666 / U1143), Centre National de la Recherche Scientifique (CNRS)-Institut Curie [Paris]-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC), Université Paris sciences et lettres, Frédéric Schmidt, and STAR, ABES
- Subjects
Bioorthogonal chemistry ,Vectorisation ,[CHIM.THER] Chemical Sciences/Medicinal Chemistry ,Immunology ,Vectorization ,Immunologie ,Chimie bioorthogonale ,Medicinal chemistry ,[CHIM.THER]Chemical Sciences/Medicinal Chemistry ,Chimie médicinale - Abstract
MAIT cells are innate-like T lymphocytes that recognize a series of microbial antigens derived from the riboflavin (vitamin B2) biosynthesis pathway, which are exclusively present in bacteria and some yeasts. The TCR dependent activation of MAIT cells requires an antigenic presentation mediated by MR1 (MHC-class I related protein) expressed mostly by antigen presenting cells (APCs). Once activated, MAIT cells can exert antimicrobial functions notably by killing pathogen-infected cells. This antimicrobial activity suggest a strong therapeutic interest in targeting MAIT cells in innovative antimicrobial immunotherapies. Unfortunately, the most active antigen discovered yet named 5-OP-RU suffers from a high chemical instability, thus making difficult the study of MAIT cell biology. 5-OP-RU is produced from a condensation reaction between its chemical precursor 5-A-RU (also unstable) and endogenous methylglyoxal. To overcome the stability issues, we designed and synthetized stable analogues of MAIT cells trying to maintain a strong potency of MAIT cells activation. We also synthetized prodrugs of 5-A-RU that were able of releasing the molecule in situ to form 5-OP-RU directly in APCs. Finally, we developed a new clickable analogue of 5-OP-RU that can be used to track and visualize MAIT cell antigens in biological tissues and cells by fluorescence microscopy., Les cellules MAIT (mucosal-associated invariant T cells) sont des lymphocytes T de l'immunité "innate-like" capables de reconnaître des antigènes microbiens dérivés de la riboflavine (vitamine B2) qui sont produits exclusivement par des bactéries et certaines levures. L'activation des cellules MAIT dépendante du TCR nécessite une présentation antigénique par la protéine MR1 (MHC-class I related protein), celle-ci étant exprimée principalement par des cellules présentatrices d'antigènes. Une fois activées, les cellules MAIT peuvent exercer des fonctions antimicrobiennes en tuant directement les cellules infectées par un pathogène. Cette activité antimicrobienne suggère un fort intérêt thérapeutique à cibler les cellules MAIT dans des immunothérapies innovantes. Malheureusement, l'antigène le plus actif des MAIT, appelé 5-OP-RU, souffre d'une importante instabilité chimique, ce qui rend l'étude des cellules MAIT difficile. Cette molécule est produite par une réaction de condensation entre son précurseur clé, le 5-A-RU (aussi instable) et du methylglyoxal endogène. Pour résoudre le problème de l'instabilité chimique, nous avons synthétisé de nouveaux analogues stables du 5-OP-RU en cherchant à maintenir une forte activité biologique sur les cellules MAIT. Nous avons également synthétisé des prodrogues du 5-A-RU capables de libérer la molécule in situ pour former ensuite du 5-OP-RU directement dans les cellules présentatrices d'antigène. Enfin, nous avons développé un analogue "clickable" du 5-OP-RU pouvant être utilisé dans le but de suivre et de visualiser les antigènes des cellules MAIT dans les tissus biologiques et les cellules par microscopie à fluorescence.
- Published
- 2020
37. Usuba, Optimizing Bitslicing Compiler
- Author
-
Mercadier, Darius, Well Honed Infrastructure Software for Programming Environments and Runtimes ( Whisper), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Sorbonne Université (France), Gilles Muller, Directeur de recherche, Inria, and Pierre-Évariste Dagand, Chargé de recherche, Sorbonne Université
- Subjects
[INFO.INFO-PL]Computer Science [cs]/Programming Languages [cs.PL] ,Vectorisation ,Vectorization ,Domain-specific language ,Cryptography ,Boolean masking ,Bitslicing ,Cryptographie ,Compilation ,Language dédié ,Masquage Booléen - Abstract
Bitslicing is a technique commonly used in cryptography to implement high-throughput parallel and constant-time symmetric primitives. However, writing, optimizing and protecting bitsliced implementations by hand are tedious tasks, requiring knowledge in cryptography, CPU microarchitectures and side-channel attacks. The resulting programs tend to be hard to maintain due to their high complexity. To overcome those issues, we propose Usuba, a high-level domain-specific language to write symmetric cryptographic primitives. Usuba allows developers to write high-level specifications of ciphers without worrying about the actual parallelization: an Usuba program is a scalar description of a cipher, from which the Usuba compiler (Usubac) automatically produces vectorized bitsliced code.When targeting high-end Intel CPUs, the Usubac applies several domain-specific optimizations, such as interleaving and custom instruction-scheduling algorithms. We are thus able to match the throughputs of hand-tuned assembly and C implementations of several widely used ciphers.Futhermore, in order to protect cryptographic implementations on embedded devices against side-channel attacks, we extend our compiler in two ways. First, we integrate into Usubac state-of-the-art techniques in higher-order masking to generate implementations that are provably secure against power-analysis attacks. Second, we implement a backend for SKIVA, a custom 32-bit CPU enabling the combination of countermeasures against power-based and timing-based leakage, as well as fault injection.; Le bitslicing est une technique utilisée pour implémenter des primitives cryptographiques efficaces et s'exécutant en temps constant. Cependant, écrire, optimiser, et sécuriser manuellement des programmes bitslicés est une tâche fastidieuse, nécessitant des connaissances en cryptographie, en microarchitecture des processeurs et en attaques par canaux cachés. Afin de remédier à ces difficultés, nous proposons Usuba, un langage dédié permettant d'implémenter des algorithmes de cryptographie symétrique. Usuba permet aux développeurs d'écrire des spécifications de haut niveau sans se soucier de leur parallélisation: un programme Usuba est une description scalaire d'une primitive, à partir de laquelle le compilateur Usuba, Usubac, produit automatiquement un code bitslicé et vectorisé.Afin de produire du code efficace pour les processeurs haut de gamme, Usubac applique plusieurs optimisations spécialement conçues pour les primitives cryptographiques, telles que l'entrelacement et l'ordonnancement d'instructions. Ainsi, le code produit par notre compilateur offre des performances comparables à du code assembleur ou C optimisé à la main.De plus, afin de générer des implémentations sécurisées contre des attaques par canaux cachés, nous proposons deux extensions de Usubac. Lorsque les attaques par analyse de courant sont un risque à considérer, Usubac est capable de protéger les implémentations qu'il produit à l'aide de masquage booléen. Si, additionellement, des attaques par injection de fautes doivent être prévenues, alors Usubac peut générer du code pour SKIVA, un processeur 32-bit offrant des instructions permettant de combiner des contre-mesures pour du code bitslicé.
- Published
- 2020
38. S aC/C formulations of the all-pairs N-body problem and their performance on SMPs and GPGPUs.
- Author
-
Šinkarovs, Artjoms, Scholz, Sven‐Bodo, Bernecky, Robert, Douma, Roeland, and Grelck, Clemens
- Subjects
MANY-body problem ,SWITCHING power supplies ,GRAPHICS processing units ,PARALLEL computers ,COMPUTER input-output equipment ,ALGORITHMS - Abstract
SUMMARY This paper describes our experience in implementing the classical N-body algorithm in S aC and analysing the runtime performance achieved on three different machines: a dual-processor 8-core Dell PowerEdge 2950 (a Beowulf cluster node, the reference machine), a quad-core hyper-threaded Intel Core-i7 based system equipped with an NVidia GTX-480 graphics accelerator and an Oracle Sparc T4-4 server with a total of 256 hardware threads. We contrast our findings with those resulting from the reference C code and a few variants of it that employ OpenMP pragmas as well as explicit vectorisation. Our experiments demonstrate that the S aC implementation successfully combines a high level of abstraction, very close to the mathematical specification, with very competitive runtimes. In fact, S aC matches or outperforms the hand-vectorised and hand-parallelised C codes on all three systems under investigation without the need for any source code modification. Furthermore, only S aC is able to effectively harness the advanced compute power of the graphics accelerator, again by mere recompilation of the same source code. Our results illustrate the benefits that S aC provides to application programmers in terms of coding productivity, source code, and performance portability among different machine architectures, as well as long-term maintainability in evolving hardware environments. Copyright © 2013 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]
- Published
- 2014
- Full Text
- View/download PDF
39. Les nanomédicaments en oncologie : nouvelles perspectives de traitement et limitations toxicologiques
- Author
-
Perez, Manon, UNIROUEN - UFR Santé (UNIROUEN UFR Santé), Université de Rouen Normandie (UNIROUEN), Normandie Université (NU)-Normandie Université (NU), and Malika Skiba
- Subjects
Vectorisation ,Nanotoxicologie ,Nanocapsules ,[SDV.SP]Life Sciences [q-bio]/Pharmaceutical sciences ,Traitement ,Nanomédicaments ,Nanovectorisation ,Cancer - Abstract
Pour pallier les limites des traitements anticancéreux classiques, diverses innovations technologiques ont émergé ces dernières années, notamment dans le domaine de la nanomédecine. Les progrès considérables des nanotechnologies permettent aujourd’hui de développer des thérapies innovantes offrant de nouvelles perspectives de traitement. La taille des nanomatériaux leur permet d’interagir à l’intérieur même des cellules et ouvre des possibilités jusqu’ici inconnues pour augmenter l’efficacité de certains traitements classiques tout en réduisant leur toxicité. Malgré les fonctions attrayantes et les perspectives d'avenir qu’ils représentent, seuls quelques nanomédicaments sont à ce jour disponibles sur le marché pour le traitement du cancer. En effet, l’application clinique de ces thérapies novatrices reste difficile et leur sécurité d’emploi est de plus en plus préoccupante. Les interactions de ces nanomédicaments avec le système immunitaire peuvent déclencher des toxicités particulières pas encore bien comprises de la communauté scientifique. L’évaluation approfondie de ces interactions est indispensable mais la littérature à ce sujet est vaste et complexe. Par conséquent, la nanotoxicologie est devenue un domaine de recherche essentiel. En raison de la complexité structurelle et chimique des nanomédicaments, les outils réglementaires classiques ne sont pas appropriés pour évaluer la sécurité d’emploi de ces formulations. La sélection de tests spécifiques est généralement indispensable et ne doit pas être négligée lors des évaluations précliniques afin d’anticiper les risques de toxicités immunologiques, et ainsi, permettre une meilleure traduction clinique à l’avenir.
- Published
- 2020
40. Evaluating the Effectiveness of a Vector-Length-Agnostic Instruction Set
- Author
-
Simon McIntosh-Smith and Andrei Poenaru
- Subjects
Vectorisation ,Data parallelism ,Computer science ,computer.software_genre ,SIMD ,Set (abstract data type) ,Instruction set ,Instruction sets ,Computer engineering ,Code (cryptography) ,Data Parallelism ,Compiler ,computer ,SVE ,Euclidean vector - Abstract
In this paper we evaluate the efficacy of the Arm Scalable Vector Extension (SVE) instruction set for HPC workloads using a set of established mini-apps. Exploiting the vector capabilities of SVE will be a key factor in achieving high performance on upcoming generations of Arm-based processors. SVE is a flexible instruction set, but its design is fundamentally different from other contemporary SIMD extensions, such as AVX or NEON, which could present a challenge to its adoption. We use a selection of mini-apps which covers a wide range of scientific application classes to investigate SVE, using a combination of static and dynamic analysis. We inspect how SVE capabilities are used in the mini- apps’ kernels, as generated by all SVE compilers available at the time of writing, for both arithmetic and memory operations. We compare our findings against similar data gathered on currently available processors. Although the extent to which vector code is generated varies by mini- app, all compilers tested successfully utilise SVE to vectorise more code than they are able to when targeting NEON, Arm’s previous-generation SIMD instruction set. For most mini-apps, we expect performance im- provements as SVE width is increased.
- Published
- 2020
- Full Text
- View/download PDF
41. L’ingénierie tissulaire comme outil pour comprendre et améliorer la délivrance de molécules thérapeutiques à l’échelle du tissu
- Author
-
Gibot, Laure, Interactions moléculaires et réactivité chimique et photochimique (IMRCP), Institut de Chimie de Toulouse (ICT), Institut de Recherche pour le Développement (IRD)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Université de Toulouse (UT)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université de Toulouse (UT)-Institut de Recherche pour le Développement (IRD)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Institut Ecologie et Environnement (INEE), Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Fluides, Energie, Réacteurs, Matériaux et Transferts (FERMAT), Institut National des Sciences Appliquées - Toulouse (INSA Toulouse), Institut National des Sciences Appliquées (INSA)-Université de Toulouse (UT)-Institut National des Sciences Appliquées (INSA)-Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université de Toulouse (UT)-Institut National des Sciences Appliquées - Toulouse (INSA Toulouse), Institut National des Sciences Appliquées (INSA)-Université de Toulouse (UT)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS), IDeAS - Interfaces Dynamiques et Assemblages Stimulables (IDeAS), Institut National des Sciences Appliquées (INSA)-Université de Toulouse (UT)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Institut de Chimie de Toulouse (ICT), Université TOULOUSE III – Paul Sabatier, Pr Nofel Merbahi, Université Toulouse III - Paul Sabatier (UT3), Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées-Centre National de la Recherche Scientifique (CNRS)-Institut de Chimie du CNRS (INC)-Institut de Chimie de Toulouse (ICT-FR 2599), Institut National Polytechnique (Toulouse) (Toulouse INP), Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD)-Université Toulouse III - Paul Sabatier (UT3), Université Fédérale Toulouse Midi-Pyrénées-Institut de Chimie du CNRS (INC)-Institut National Polytechnique (Toulouse) (Toulouse INP), and Université Fédérale Toulouse Midi-Pyrénées-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD)
- Subjects
electroporation ,skin ,nanovecteurs polymères ,[SDV.BIO]Life Sciences [q-bio]/Biotechnology ,extracellular matrix ,matrice extracellulaire ,[SDV.CAN]Life Sciences [q-bio]/Cancer ,[SDV.BC]Life Sciences [q-bio]/Cellular Biology ,microvascularization ,Ingénierie tissulaire ,MESH: Techniques et équipements analytiques, diagnostiques et thérapeutiques [E] ,vectorization ,sphéroïde ,électroporation ,soft matter ,vectorisation ,polymer nanovectors ,matière molle ,thérapie photodynamique ,[CHIM.POLY]Chemical Sciences/Polymers ,photodynamic therapy ,microvascularisation ,spheroid ,peau ,tissue engineering ,drug delivery - Abstract
Pour assurer l’efficacité thérapeutique d’une molécule-médicament, il faut prendre en compte non seulement son activité intrinsèque, mais aussi les nombreuses barrières physiques, chimiques et biologiques qu’elle rencontrera de son site d’injection jusqu’à sa cible et qui peuvent limiter son action. Ces barrières sont aussi diverses que le passage de l’endothélium, des jonctions intercellulaires, la membrane plasmique, la diffusion à travers la matrice extracellulaire, les zones hypoxiques et de faible pH dans les tumeurs… Dans le contexte du cancer, en associant les médicaments avec un système d'administration adéquat, aussi appelé vecteur, celui-ci va améliorer la pharmacocinétique de la molécule active réduisant sa toxicité, les effets secondaires pour les cellules saines et augmenter sa biodistribution au niveau tumoral. Ce concept de transport de la molécule-médicament vers le tissu cible s’appelle la vectorisation (drug delivery en anglais). Dans ce document nous aborderons de façon plus approfondie deux types de vectorisation pour délivrer des molécules-médicament : 1) La vectorisation par des méthodes physiques telles que l’application d’un champ électrique externe, appelée électroporation et 2) La vectorisation par des vecteurs chimiques via des nanoparticules polymères de types micelles ou polymersomes dans le cadre de la thérapie photodynamique (PDT).Le nombre incroyable d’essais cliniques en cancérologie qui ont échoué malgré des résultats extrêmement prometteurs obtenus avec des modèles animaux in vivo ou des cellules in vitro montrent la nécessité de développer d'autres approches translationnelles pour enrichir le lien entre les études in vitro et les applications cliniques. C’est ce que propose l’ingénierie tissulaire qui vise à produire en laboratoire à partir de co-cultures de cellules humaines des tissus humains en trois dimensions riches en matrice extracellulaire. L’objectif des travaux présentés dans ce document est de mieux comprendre les mécanismes et l’efficacité de transfert des molécules-médicaments par électroporation et nanovecteurs polymères à l’échelle tissulaire dans un contexte cutané et/ou tumoral. Les résultats montrent sans surprise que la vectorisation améliore l’efficacité des molécules-médicament dans un contexte tissulaire 3D. Le point original qui est soulevé par les résultats et la discussion est l’importance de s’intéresser à l’effet intrinsèque du vecteur (per se), à savoir le champ électrique seul ou les polymères. Nous avons ainsi démontré par exemple que le champ électrique, un stimulus purement physique, induit la transdifférenciation de fibroblastes en myofibroblastes et le remodelage des collagènes cutanés via l’augmentation de la quantité de métalloprotéinases (MMPs) actives.En conclusion, l’ingénierie tissulaire est une approche attractive pour étudier la délivrance de molécules-médicament à l’échelle tissulaire et donc optimiser leur efficacité thérapeutique, tout en limitant le recours aux expérimentations animales en accord avec la politique éthique des 3R.
- Published
- 2020
42. Optimisation of the two-dimensional hydraulic model LISFOOD-FP for CPU architecture
- Author
-
Paul D. Bates, Christopher C. Sampson, Jeffrey Neal, Toby Dunne, and Andrew Smith
- Subjects
Environmental Engineering ,Vectorisation ,010504 meteorology & atmospheric sciences ,Flood myth ,Computer science ,Ecological Modeling ,Computation ,0208 environmental biotechnology ,Sample (statistics) ,Context (language use) ,02 engineering and technology ,Parallel computing ,Solver ,01 natural sciences ,020801 environmental engineering ,Constraint (information theory) ,Test case ,HPC ,LISFLOOD-FP ,Parallelisation ,Single-core ,Flood inundation modelling ,Software ,0105 earth and related environmental sciences - Abstract
Flood inundation models are increasingly used for a wide variety of river and coastal management applications. Nevertheless, the computational effort to run these models remains a substantial constraint on their application. In this study four developments to the LISFLOOD-FP 2D flood inundation model have been documented that: 1) refine the parallelisation of the model; 2) reduce the computational burden of dry cells; 3) reduce the data movements between CPU and RAM; and 4) vectorise the core numerical solver. The value of each of these developments in terms of compute time and parallel efficiency was tested on 12 test cases. For realistic test cases, improvements in single core performance of between 4.2x and 8.4x were achieved, which when combined with parallelisation on 16 cores resulted in computation times 34-60x shorter than previous LISFLOOD-FP models on one core. Results were compared to a sample of commercial models for context.
- Published
- 2018
- Full Text
- View/download PDF
43. Brian hears: online auditory processing using vectorisation over channels
- Author
-
Bertrand eFontaine, Dan F. M Goodman, Victor eBenichoux, and Romain eBrette
- Subjects
brian ,gpu ,python ,auditory filter ,vectorisation ,Neurosciences. Biological psychiatry. Neuropsychiatry ,RC321-571 - Abstract
The human cochlea includes about 3000 inner hair cells which filter sounds at frequencies between 20 Hz and 20 kHz. This massively parallel frequency analysis is reflected in models of auditory processing, which are often based on banks of filters. However, existing implementations do not exploit this parallelism. Here we propose algorithms to simulate these models by vectorising computation over frequency channels, which are implemented in Brian Hears, a library for the spiking neural network simulator package Brian. This approach allows us to use high-level programming languages such as Python, as the cost of interpretation becomes negligible. This makes it possible to define and simulate complex models in a simple way, while all previous implementations were model-specific. In addition, we show that these algorithms can be naturally parallelised using graphics processing units, yielding substantial speed improvements. We demonstrate these algorithms with several state-of-the-art cochlear models, and show that they compare favorably with existing, less flexible, implementations.
- Published
- 2011
- Full Text
- View/download PDF
44. New method for post-processing crosslinking reaction
- Author
-
Abbas, L., Bouquey, M., Flat, J.J., and Muller, R.
- Subjects
- *
POLYMERS , *CROSSLINKING (Polymerization) , *MALEIC anhydride , *ETHYLENE , *COPOLYMERS - Abstract
Abstract: A new processing method to process functionalized polymer materials that are able to crosslink has been developed. The crosslinking reaction occurs after the melt processing stage at moderate temperature and without the addition of external reagents. Maleic anhydride-ethylene copolymer was processed with 1,4-butanediol in the presence of para-toluene sulfonic acid as catalyst. To prevent the reaction in the melt, it was chosen to trap the 1,4-butanediol onto vector particles. The extent of the crosslinking reaction during the mixing at 110°C has been studied for three different particles taken as vector particles. It was found that Orgasol® polyamide particles are efficient to inhibit the crosslinking reaction in the mixer. The reaction then takes place at 40°C and after 100h the cured samples appeared to be highly and homogeneously crosslinked with a gel content above 90%. [Copyright &y& Elsevier]
- Published
- 2008
- Full Text
- View/download PDF
45. Functional amphiphilic and biodegradable copolymers for intravenous vectorisation
- Author
-
Van Butsele, K., Jérôme, R., and Jérôme, C.
- Subjects
- *
CANCER treatment , *POLYETHYLENE oxide , *BLOOD flow , *TUMORS , *POLYMERIC drug delivery systems - Abstract
Abstract: This paper aims at reporting on the design of polymeric drug nanocarriers used in cancer therapy, with a special emphasis on the control of their biodistribution. First, the prominent role of poly(ethylene oxide) in the lifetime of nanocarriers circulating in the blood stream is highlighted, and the origin of a passive targeting based on a difference in the anatomy of tumors and normal tissues is discussed. The main body of the review is devoted to the targeting of nanocarriers towards tumors and the underlying concepts. As a rule, either the constitutive polymer is stimuli-responsive and the locus of drug release is where the stimulation occurs, or a ligand endowed with specific recognition is grafted onto the nanocarrier. Finally, the fate of the nanocarrier after drug delivery and the bioelimination of the polymer(s) involved are briefly considered. [Copyright &y& Elsevier]
- Published
- 2007
- Full Text
- View/download PDF
46. Reversible vectorisation of 3D digital planar curves and applications
- Author
-
Sivignon, Isabelle, Dupont, Florent, and Chassery, Jean-Marc
- Subjects
- *
ALGORITHMS , *DIGITAL images , *COMPUTER simulation , *PLANE curves , *POLYGONS , *DIGITAL image processing , *DIGITIZATION - Abstract
This paper tackles the problem of the computation of a planar polygonal curve from a digital planar curve, such that the digital data can be exactly retrieved from the polygonal curve. The proposed transformation also provides an analytical modelling of a digital plane segment as a discrete polygon composed of a face, edges and vertices. A dual space representation of lines and planes is used to ensure that the computed curve remains inside the digital curve, and this tool enables to define a very efficient algorithm. Applied on the digital plane segments resulting from the decomposition of a digital surface, this algorithm provides a set of polygons modelling exactly the digital surface. [Copyright &y& Elsevier]
- Published
- 2007
- Full Text
- View/download PDF
47. Synthesis of versatile chemical tools toward a structure/properties relationships study onto targeting colloids
- Author
-
Jolimaitre, Pascale, Poirier, Cécile, Richard, Antoine, Blanpain, Annick, Delord, Brigitte, Roux, Didier, and Bourel-Bonnet, Line
- Subjects
- *
DRUG delivery systems , *ALDEHYDES , *COLLOIDS , *AMORPHOUS substances , *SOLUTION (Chemistry) , *SURFACE chemistry , *ORGANIC compounds , *PEPTIDES - Abstract
Abstract: As part of a drug delivery project, four aldehydes of the type Pam-Lys(Pam)-spacer-CO-CHO were synthesized to be included in targeting colloids. Though amphiphilic, they were obtained within reasonable yields (18–55%) and with high RP-HPLC purity (∼90%). Parallely, six complementary targeting peptides of the type H2N-NH-CH2-CO-spacer-YGRGDSP-NH2 were prepared to be anchored onto colloids. Isolated yields are related to the spacer length and nature. To easily and rapidly modulate the distance between the peptide and the vesicle, every partners were elaborated on solid phase and the expected constructions were obtained by hydrazone ligation. One possible application is presented here with multilamellar vesicles targeting HUVEC cells. Preliminary results prove that the fine-tuning of the spacer length permits to optimize the recognition toward the target cells. [Copyright &y& Elsevier]
- Published
- 2007
- Full Text
- View/download PDF
48. Teseo: A vectoriser of historical seismograms
- Author
-
Pintore, Stefano, Quintiliani, Matteo, and Franceschi, Diego
- Subjects
- *
SEISMOGRAMS , *SEISMOMETRY , *GEOGRAPHIC information systems , *ENVIRONMENTAL impact analysis - Abstract
Abstract: Historical seismograms contain a rich harvest of information useful for the study of past earthquakes. It is necessary to extract this information by digitising the analogue records if modern analysis is required. Teseo has been developed for quick and accurate digitisation of seismogram traces from raster files, introducing a vectorisation step based on piecewise cubic Bézier curves. The vectoriser can handle greyscale images stored in a suitable file format and it offers three concurrent vectorisation methods: manual, automatic by colour selection, and automatic by neural networks. The software that implements the methods described is distributed with open source license. [Copyright &y& Elsevier]
- Published
- 2005
- Full Text
- View/download PDF
49. Regioselective synthesis of peptidic derivatives and glycolamidic esters of Methotrexate
- Author
-
Castex, Cédric, Lalanne, Christophe, Mouchet, Patrick, Lemaire, Marc, and Lahana, Roger
- Subjects
- *
METHOTREXATE , *PEPTIDES , *ANTINEOPLASTIC agents , *ORGANIC compounds - Abstract
Abstract: Convenient methods for regioselective syntheses of Methotrexate peptidic conjugates are described. Solid phase synthesis for derivatives of Methotrexate containing an amide bond has been applied and showed higher efficiency than liquid phase synthesis. Synthetic pathways for regioselective preparation of Glycolamidic ester derivatives of Methotrexate were also developed using 4-amino-4-deoxy-N10-methylpteroic acid as starting material. These ester bonds were obtained either in solution or by using a combined liquid and solid phase synthesis. [Copyright &y& Elsevier]
- Published
- 2005
- Full Text
- View/download PDF
50. Second-order bilinear calibration: the effects of vectorising the data matrices of the calibration set
- Author
-
Faber, Nicolaas (Klaas) M., Ferré, Joan, Boqué, Ricard, and Kalivas, John H.
- Subjects
- *
PREDICTION theory , *DATA analysis , *LEAST squares - Abstract
In a groundbreaking paper, Linder and Sundberg [Chemometr. Intell. Lab. Syst. 42 (1998) 159] developed a statistical framework for the calibration of second-order bilinear data. Within this framework, they formulated three different predictor construction methods [J. Chemom. 16 (2002) 12], namely the so-called naı¨ve method, the bilinear least squares (BLLS) method, and a refined version of the latter that takes account of the calibration uncertainty. Elsewhere [J. Chemom. 15 (2001) 743], a close relationship is established between the naı¨ve method and the generalized rank annihilation method (GRAM) by comparing expressions for prediction variance. Here it is proved that the BLLS method can be interpreted to work with vectorised data matrices, which establishes an algebraic relationship with so-called unfold partial least squares (PLS) and unfold principal component regression (PCR). It is detailed how these results enable quantifying the effects of vectorising bilinear second-order data matrices on analytical figures of merit and variance inflation factors. [Copyright &y& Elsevier]
- Published
- 2002
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.