Search

In recent years, some strategies (e.g., server consolidation by means of virtualisation techniques) helped the managers of large Information Technology (IT) infrastructures to limit, when possible, the use of hardware resources in order to provide reliable services and to reduce the Total Cost of Ownership (TCO) of such infrastructures. Moreover, with the advent of Cloud computing, a resource usage rationalisation can be pursued also for the users applications, if this is compatible with the Quality of Service (QoS) which must be guaranteed. In this perspective, modern datacenters are “elastic”, i.e., able to shrink or enlarge the number of local physical or virtual resources from private/public Clouds. Moreover, many of large computing environments are integrated in distributed computing environment as the grid and cloud infrastructures. In this document, we report some advances in the realisation of a utility, we named Adaptive Scheduling Controller (ASC) which, interacting with the datacenter resource manager, allows an effective and efficient usage of resources, also by means of users jobs classification. Here, we focus both on some data mining algorithms which allows to classify the users activity and on the mathematical formalisation of the functional used by ASC to find the most suitable configuration for the datacenter’s resource manager. The presented case study concerns the SCoPE infrastructure, which has a twofold role: local computing resources provider for the University of Naples Federico II and remote resources provider for both the Italian Grid Infrastructure (IGI) and the European Grid Infrastructure (EGI) Federated Cloud.

We explore the filtering properties of wavelets functions in order to develop accurate and efficient numerical algorithms for Image Restoration problems. We propose a parallel implementation for MIMD distributed memory environments. The key insight of our approach is the use of distributed versions of Level 3 Basic Linear Algebra Subprograms as computational building blocks and the use of Basic Linear Algebra Communication Subprograms as communication building blocks for advanced architecture computers. The use of these low-level mathematical software libraries guarantees the development of efficient, portable and scalable high-level algorithms and hides many details of the parallelism from the user's point of view. Numerical experiments on a simulated image restoration applications are shown. The parallel software has been tested on a 12 nodes IBM SP2 available at the Center for Research on Parallel Computing and Supercomputers in Naples (Italy).

In the last decades, diagnosing medical images has heavily relied on digital imaging. As a consequence, huge amounts of data produced by modern medical instruments need to be processed, organized, and visualized in a suitable response time. Many efforts have been devoted to the development of digital Picture Archiving and Communications Systems (PACS) which archive and distribute image information across a hospital and provide web accessto avoid the expensive deployment of a large number of such systems. On the other hand, this approach does non solve problems related to the increasing demand of high performance computing and storage facilities, which cannot be placed within a hospital. In this work we describe MedIGrid, an application that enables nuclear doctors to transparently use high performance computers and storage systems for the PET/SPECT image processing, management, visualization and analysis. MedIGrid is the result of the joint efforts of a group of researchers committed to the development of a distributed application to test and deploy new reconstruction methods in clinical environments. Th outcomes of this work include a set of platform indipendent software tools to read medical images, control the execution of computing intensive tomographic algorithms, and explore the reconstructed tomographic volumes. We also describe how the collaboration among different research groups has contributed to the integration of the application into a single framework.

Data Assimilation (DA) is an uncertainty quantification technique used to incorporate observed data into a prediction model in order to improve numerical forecasted results. As a crucial point into DA models is the ill conditioning of the covariance matrices involved, it is mandatory to introduce, in a DA software, preconditioning methods. Here we present first results obtained introducing two different preconditioning methods in a DA software we are developing (we named S3DVAR) which implements a Scalable Three Dimensional Variational Data Assimilation model for assimilating sea surface temperature (SST) values collected into the Caspian Sea by using the Regional Ocean Modeling System (ROMS) with observations provided by the Group of High resolution sea surface temperature (GHRSST). We present the algorithmic strategies we employ and the numerical issues on data collected in two of the months which present the most significant variability in water temperature: August and March.

We develop innovative algorithms for solving the strong-constraint formulation of four-dimensional variational data assimilation in large-scale applications. We present a space-time decomposition approach that employs domain decomposition along both the spatial and temporal directions in the overlapping case and involves partitioning of both the solution and the operators. Starting from the global functional defined on the entire domain, we obtain a type of regularized local functionals on the set of subdomains providing the order reduction of both the predictive and the data assimilation models. We analyze the algorithm convergence and its performance in terms of reduction of time complexity and algorithmic scalability. The numerical experiments are carried out on the shallow water equation on the sphere according to the setup available at the Ocean Synthesis/Reanalysis Directory provided by Hamburg University., Comment: Received: 10 March 2020 / Revised: 29 November 2021 / Accepted: 7 March 2022

Large-scale problems in engineering and science often require the solution of sparse linear algebra problems and the Krylov subspace iteration methods (KM) have led to a major change in how users deal with them. But, for these solvers to use extreme- scale hardware efficiently a lot of work was spent to redesign both the KM algorithms and their implementations to address challenges like extreme concurrency, complex memory hierarchies, costly data movement, and heterogeneous node architectures. All the redesign approaches bases the KM algorithm on block-based strategies which lead to the Block-KM (BKM) algorithm which has high granularity (i.e., the ratio of computation time to communication time). The work proposes novel parallel revisitation of the modules used in BKM which are based on the overlapping of communication and computation. Such revisitation is evaluated by a model of their granularity and verified on the basis of a case study related to a classical problem from numerical linear algebra.

We herein describe the performance evaluation of a modular implementation of the MGRIT (MultiGrid-In-Time) algorithm within the context of the PETSc (the Portable, Extensible Toolkit for Scientific computing) library. Our aim is to give the PETSc users the opportunity of testing the MGRIT parallel-in-time approach as an alternative to the Time Stepping integrator (TS), when solving their problems arising from the discretization of linear evolutionary models. To this end, we analyzed the performance parameters of the algorithm in order to underline the relationship between the configuration factors and problem characteristics, intentionally overlooking any accuracy issue and spacial parallelism.

The deployment, management and total cost of ownership of large computing environments always involve huge investments. These systems, once in production, have to meet the needs of users belonging to large and heterogeneous communities whose demands are often conflicting. Only an efficient and effective use of these systems can pay back the investment carried out. The infrastructure needs to be sustained also by means of the building of an "appropriate" community of users (the "catchment area"), to be, firstly, recruited and then driven to a conscious use of resources. Based on the experience we gained at the University of Naples Federico II, in the activities related to "support users and applications" on the SCoPE system, we "enjoyed" to write our "recipe". Since 2008, also in the framework of ReCaS, that recipe represents our way to recruit new users and to consolidate the "catchment area" in terms of both number and awareness in the use of the infrastructure.

The minimisation of the total cost of ownership is hard to be faced by the owners of large scale computing systems, without affecting negatively the quality of service for the users. Modern datacenters, often included in distributed environments, appear to be ``elastic'', i.e., they are able to shrink or enlarge the number of local physical or virtual resources, also by recruiting them from private/public clouds. This increases the degree of dynamicity, making the infrastructure management more and more complex. Here, we report some advances in the realisation of an adaptive scheduling controller (ASC) which, by interacting with the datacenter resource manager, allows an effective and an efficient usage of resources. In particular, we focus on the mathematical formalisation of the ASC's kernel that allows to dynamically configure, in a suitable way, the datacenter resources manager. The described formalisation is based on a probabilistic approach that, starting from both a hystorical resources usage and on the actual users request of the datacenter resources, identifies a suitable probability distribution for queue time with the aim to perform a short term forecasting. The case study is the SCoPE datacenter at the University of Naples Federico II.

Cloud computing has been widely adopted as an alternative paradigm with respect to Grid computing to provide resources because of the advantages it brings to both users and providers. GaaS (Grid as a Service) was born with the aim to enable the use of Cloud-provided resources with well-established Grid-like interfaces, avoiding resources to learn new resources access and use models. in previous experiences, we developed a prototype for GaaS, by using the most used Cloud open source technologies and frameworks available at that time (e.g., Eucalyptus, OpenNebula). here, we present the final stage of our work on GaaS that is now based on several de facto and consolidated standards in Cloud, Virtualization and Grid worlds: OpenStack, KVM and EMI. Furthermore, we present both i) a production level of GaaS and ii) GApp (GaaS APPlication), mobile interface to the GaaS framework.

L'iniziativa SCoPE@Scuola nasce con l'obiettivo principale di stimolare la curiosità degli studenti delle scuole superiori sulle tematiche del supercalcolo quale strumento indispensabile per la soluzione dei problemi-sfida della scienza e della tecnologia. Le tematiche connesse con il supercalcolo coinvolgono, infatti, un ampio spazio di conoscenze e competenze: dalla modellizzazione matematica dei problemi alla progettazione di algoritmi, dall'implementazione del software alla progettazione e gestione di sistemi informatici complessi. L'iniziativa, partita alla fine del 2014, ha coinvolto tre scuole, è durata sei mesi (articolandosi in più incontri e tipi di attività) e si è conclusa, per questa prima edizione, alla fine del mese di maggio 2015.

A key issue in designing application codes which are able to effectively take advantage of the available high performance computing resources is to deal with the software management and development complexity. This is even more true for imaging applications. This work is the first dowel in a rather wide mosaic, aimed at construction of a PSE (Problem Solving Environment) oriented to imaging applications. We discuss computational efforts towards the development of a distributed software environment, enabling the transparent use of high performance computers and storage systems for 3-D Echocardiographic sequences denoising via nonlinear diffusion filtering. More precisely, we describe a component-based approach for the development of an integrated software environment relying on the Portable, Extensible Toolkit for Scientific Computation (PETSc). Our approach uses a distributed memory model where we hide, within the PETSc parallel objects, the details of internode communications while intranode communications have been handled at a higher level. We report some experiences we made on an in vivo acquired 3-D Echocardiographic sequence obtained by means of a rotational acquisition technique using Tomtec Imaging system. btained by means of a rotational acquisition technique using Tomtec Imaging system.

The development of scientific software, reliable and efficient, in distributed computing environments, requires the identification and the analysis of issues related to the design and the deployment of algorithms for high-performance computing architectures and their integration in distributed contexts. In these environments, indeed, resources efficiency and availability can change unexpectedly because of overloading or failure i.e. of both computing nodes and interconnection network. The scenario described above, requires the design of mechanisms enabling the software to "survive" to such unexpected events by ensuring, at the same time, an effective use of the computing resources. Although many researchers are working on these problems for years, fault tolerance, for some classes of applications is an open matter still today. Here we focus on the design and the deployment of a checkpointing/migration system to enable fault tolerance in parallel applications running in distributed environments. In particular we describe details about HADAB, a new hybrid checkpointing strategy, and its deployment in a meaningful case study: the PETSc Conjugate Gradient algortithm implementation. The related testing phase has been performed on the University of Naples distributed infrastructure (S.Co.P.E. infrastructure).

In this paper we are concerned with improvements and enhancements of a medical imaging grid-enabled infrastructure, named MedIGrid, oriented to the transparent use of resource-intensive applications for managing, processing and visualizing biomedical images. We describe an implementation of the MedIGrid PSE in an LCG/gLite environment. Wepsilall mainly focus on how to exploit the features of the new middleware environment to improve the efficiency and the services reliability of the PSE; further, some comments will be devoted to how to modify, extend and/or improve the underlying numerical components.

Clinical diagnosis environments often require the availability of processed data in real-time, unfortunately, reconstruction times are prohibitive on conventional computers, neither the adoption of expensive parallel computers seems to be a viable solution. Here, we focus on development of mathematical software on high performance architectures for Total Variation based regularization reconstruction of 3D SPECT images. The software exploits the low-cost of Beowulf parallel architectures. The activity has been developed within the Project: “Inverse Problems in Medical Imaging” partially supported by MURST, grant number MM01111258