Search

Software Defined Radio (SDR) platforms are valuable for research and development activities or high-end systems that demand real-time adaptable wireless protocols. While low latency can be achieved using the dedicated digital processing unit of a state-of-the-art SDR platform, its Radio Frequency (RF) front-end often poses a limitation in terms of turnaround time (TT), the time needed for switching from the receiving to the transmitting mode (or vice versa). Zero Intermediate Frequency (ZIF) transceivers are favorable for SDR, but suffer from self-interference even if the device is not currently transmitting. The strict MAC-layer requirements of Time Division Duplex (TDD) protocols like Wi-Fi cannot be achieved using configurable ZIF transceivers without having to compromise receiver sensitivity. Using a novel approach, we show that the TT using the AD9361 RF front-end can be as low as 640 ns, while the self-interference is at the same level as achieved by the conventional TDD mode, which has a TT of at least 55 {\mu}s. As compared to Frequency Division Duplex (FDD) mode, a decrease of receiver noise floor by about 13 dB in the 2.4 GHz band and by about 4.5 dB in the 5 GHz band is achieved., Comment: Accepted by IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM 2022) as Work-In-Progress paper

Fifth Generation (5G) systems are envisaged to support a wide range of applications scenarios with varying requirements. 5G architecture includes network slicing abilities which facilitate the partitioning of a single network infrastructure on to multiple logical networks, each tailored to a given use case, providing appropriate isolation and Quality of Service (QoS) characteristics. Radio Access Network (RAN) slicing is key to ensuring appropriate QoS over multiple domains; achieved via the configuration of multiple RAN behaviours over a common pool of radio resources. This Paper proposes a novel solution for efficient resource allocation and assignment among a variety of heterogeneous services, to utilize the resources while ensuring maximum QoS for network services. First, this paper evaluates the effectiveness of different wireless data bearers. Secondly, the paper proposes a novel dynamic resource allocation algorithm for RAN slicing within 5G New Radio (NR) networks utilising cooperative game theory combined with priority-based bargaining. The impact of this work to industry is to provide a new technique for resource allocation that utilizes cooperative bargaining to ensure all network services achieve minimum QoS requirements – while using application priority to reduce data transfer time for key services to facilitate decreased turnaround time at the gate.

Ordering service in a permissioned blockchain platform like Hyperledger Fabric groups the transactions into various blocks. These blocks are linked together via a hash chain to provide immutability. Transactions with less block height get validated before the transactions with more block height. Within a single block, the transactions get validated in the same order they are arranged inside the block. This structure imposes a total order among transactions. This total ordering ensures that all the peer nodes reach the same final state once all the transactions are validated. This total ordering prevents parallel validation of transactions. A given transaction may not be dependent on all the transactions preceding it. However, such a transaction still cannot be validated before completion of validation for all the preceding transactions. This paper presents a blockchain system to support partial ordering among non-dependent transactions. The proposed blockchain system aims to achieve high transaction throughput by enabling parallel transaction validation and reducing the transaction turnaround time by eliminating the need for the transactions to wait for non-dependent predecessor transactions. The proposed blockchain system ensures that all peer nodes reach the same final state even if the order in which they validate the transactions is different.

Change is one of the biggest contributors to service outages. With more enterprises migrating their applications to cloud and using automated build and deployment the volume and rate of changes has significantly increased. Furthermore, microservice-based architectures have reduced the turnaround time for changes and increased the dependency between services. All of the above make it impossible for the Site Reliability Engineers (SREs) to use the traditional methods of manual risk assessment for changes. In order to mitigate change-induced service failures and ensure continuous improvement for cloud native services, it is critical to have an automated system for assessing the risk of change deployments. In this paper, we present an AI-based system for proactively assessing the risk associated with deployment of application changes in cloud operations. The risk assessment is accompanied with actionable risk explainability. We discuss the usage of this system in two primary scenarios of automated and manual deployment. In automated deployment scenario, our approach is able to alert SREs on 70 % of problematic changes by blocking only 1.5 % of total changes and recommending human intervention. In manual deployment scenario, our approach recommends the SREs to perform extra due diligence for 2.8 % of total changes to capture 84 % of problematic changes.

Job scheduling aims to minimize the turnaround time on the submitted jobs while catering to the resource constraints of High Performance Computing (HPC) systems. The challenge with scheduling is that it must honor job requirements and priorities while actual job run times are unknown. Although approaches have been proposed that use classification techniques or machine learning to predict job run times for scheduling purposes, these approaches do not provide a technique for reducing underprediction, which has a negative impact on scheduling quality. A common cause of underprediction is that the distribution of the duration for a job class is multimodal, causing the average job duration to fall below the expected duration of longer jobs. In this work, we propose the Top Percent predictor, which uses a hierarchical classification scheme to provide better accuracy for job run time predictions than the user-requested time. Our predictor addresses multimodal job distributions by making a prediction that is higher than a specified percentage of the observed job run times. We integrate the Top Percent predictor into scheduling algorithms and evaluate the performance using schedule quality metrics found in literature. To accommodate the user policies of HPC systems, we propose priority metrics that account for job flow time, job resource requirements, and job priority. The experiments demonstrate that the Top Percent predictor outperforms the related approaches when evaluated using our proposed priority metrics.

With the prevalence of machine learning applications, an increasing number of machine learning tasks is transplanted into cloud computing platform. The cloud for machine learning consists of heterogeneous resources such as GPUs, CPUs, memory, etc, which brings challenges for resource management. So, how to schedule machine learning tasks and allocate appropriate GPU resources for computing, so that the cluster can maximize the use of resources and reduce task computing time has become a concern in the industry and academia. This paper proposes a scheduling strategy named Proteus, which is based on Lyapunov optimization. Through the Proteus, we can make our tasks have a minimum turnaround time in a long time sequence, which is the time from the submission of the task to the completion of the task. By performing a comprehensive analysis, we implement the scheduling algorithm and conducts several simulation experiments. The experimental result shows that our scheduling strategy can achieve significant results in most task scheduling environments, reducing the turnaround time of tasks to 40%–50% of the original. This shows that the Proteus can provide higher resource utilization and performance of cloud clusters and reduce task turnaround time.

An automatic optimization methodology based on AI algorithm is proposed to achieve multi-targeting of various devices in 0.13 μm next BCD process development. The optimized process conditions are simultaneously provided with satisfying various ET-specs of the BCD devices from our method and TCAD analysis. The method has practically been applied to well ion implantation processes shared with seven different devices, and its targeting rate of 87% has been verified through silicon evaluation. Its turnaround time (TAT) is reduced by 90% compared to conventional procedure.

In this ever growing, competitive market with Technological advancements Reuse management plays an important role in reducing the time to market and optimizing the R&D to sales ratio. It’s now clear that to achieve better software, more quickly and at lower cost, we need a design process that is based on systematic software reuse In any R&D engineering center like ours, the reuse management strives to launch competitive, focus on innovative products with high profiting business models in the shortest lead time possible. In this paper, a novel approach is proposed that embarks the journey of reuse component identification wrt,•Common across customer projects,•Common across product lines•Common across Business unit levelThis method is derived from both AutoSAR and Non AutoSAR project architecture. This method has been piloted in our Engineering Center and has proven to be a very method for computing (evaluating) and thereby improving the reuse. This in turn minimizes effort and cost in product development and adapting market strategy in its business models. Keywords-Feature, Reuse component identification, Reuse Index calculation, Market estimation, Value addition

Deployment of Deep Neural Networks (DNNs) on edge devices is highly desirable to address user privacy concerns and minimize the turnaround time of AI applications. However, the execution of DNN models on a battery-operated device requires a highly optimized implementation specific to the target hardware. Moreover, as different layers of a DNN exhibit distinct computation and memory characteristics, it is imperative to optimize each layer separately. This is in contrast to the widely deployed library-based approach where all the configurations of DNN operations share the same implementation. In this paper, we address this issue by auto-tuning the implementation of Long Short Term Memory (LSTM) operations which are widely used in sequence based AI applications. To exhaustively search through the space of optimizations and its parameters, we develop a high-level autotuning framework based on Halide. We use grid search to find the parameters that lead to minimum runtime and further present TPE based search method to find the near-optimal runtime in a limited number of trials. We observe 2.2× —3.1× speedup in execution time for LSTM layers used in widely deployed GNMT and DeepSpeech2 models.

Vehicle inspection systems (VIS) have been extensively used by automobile manufacturers, maintenance companies, and the traffic administrative departments for the inspection of safety, reliability, and other indicators of both newly produced and in-used vehicles. How to increase their productivity by effectively operating such systems is an interesting and crucial problem. Because of the space limitation of testing field in such a system, a vehicle must complete pre-set inspection items at multiple stations in sequence. Hence, it is very challenging to sequence the vehicles to be tested to shorten the unnecessary waiting time for vehicle testing. This paper investigates the vehicle scheduling problem of such systems, where the time for vehicle movement in the system can be ignored. The system is modeled by resource-oriented Petri net, a graphical and mathematical modeling tool. Based on the model, this work analyzes the testing cycle time and develops a heuristic algorithm to efficiently solve the vehicle scheduling problem in such a system. For the first time, a heuristic algorithm for generating an optimal vehicle queue for testing is proposed. Industrial examples are used to illustrate the proposed algorithm and results show that a significant reduction in total test turnaround time can be obtained in comparison with the existing methods.

The cluster system presents multi-level and complex heterogeneous features with GPUs computing. Job scheduling technology determines computing resource utilization and throughput of the cluster system. On the basis of heterogeneous cluster system, the two-level scheduling technology is proposed in this paper. In the first level, a job priority scheduling algorithm based on user value is designed at the cluster level using analytical hierarchy processes. The second level of scheduling is to act on the computing nodes that contain GPUs, and design a scheduling algorithm based on GPU sharing policy that manages GPU jobs in a more fine-grained manner. Through experiments, it shows that the job priority scheduling algorithm based on user value can ensure that jobs with high user value are executed first, and fairness is guaranteed. The scheduling algorithm based on GPU sharing technology can significantly improve the utilization of GPU and video memory, reducing the average waiting time of jobs by 55.54% and the average turnaround time of jobs by 28.83%.

In this paper, we consider the server allocation problem in edge computing. We consider a system model where there are a number of areas or locations, each of which has an associated Base Station (BS), where we can deploy an edge cloud with multiple servers. Each edge cloud will process application requests received at the corresponding BS from users in the corresponding area. The system manager/operator has a budget to deploy a given number of servers to the BSs. Our goal is to come up with a server allocation plan, i.e., how many servers to deploy at each of the BSs, such that the overall average turnaround time of application requests generated by all the users is minimized. In order to achieve the optimal solution for the problem, we resort to queueing theory and model each edge cloud as an M/M/c queue. Analysis on the problem motivates a Largest Weighted Reduction Time First (LWRTF) algorithm to assign servers to edge clouds. Numerical comparisons among various algorithms verify that Algorithm LWRTF has near-optimal performances in terms of minimizing the average turnaround time. Simulation results using the CloudSim Plus simulation tool also verify that Algorithm LWRTF achieves better performances compared to other reasonably designed heuristic algorithms.

Listen-Before-Talk (LBT) is an essential function of many MAC protocols and a key mechanism of sharing spectrum in an uncoordinated manner. However, this simple but effective concept remains a major challenge for Software-Defined Radios (SDRs). If performing the protocol stack on a host PC, due to their structure, SDRs have inherent latencies built in. These latencies increase their reaction time, the so-called turnaround time, significantly, compared to their conventional radio counterparts. Unfortunately, this means that they cannot comply with the LBT channel access procedures used in modern protocols like IEEE 802.11 or LTE Licensed-Assisted Access (LTE-LAA). Given the flexibility and rapid-prototyping capabilities of SDRs, these protocols could clearly benefit from such SDR-based implementations. In this paper, we fill this gap and present a design approach for SDRs supporting LBT. We particularly focus on the rather complicated timing issues. As a proof-of-concept, we showcase our LBT-enabled implementation of the srsLTE software stack for LTE.

The article presents the results of analyzing the failure statistics of the main components of steam turbines of thermal power plants. Based on the results of the analysis established that the complexity of technological processes, taking into account turnaround time, failure rate, the production cycle of the equipment residual life allows to develop approaches to assessing the health of each element with the aim of obtaining reliable and representative sample of failure statistics to reliability assessment of steam turbine plants. It is shown that the statistics of finding equipment in a state of partial failure is an important factor for improving the overall reliability of equipment. The results obtained can be used to create predictive models that provide approaches to prolonging the operational state of elements of steam turbine installations, as well as for implementing projects of digital energy systems that include subsystems for monitoring and diagnostics of power equipment of thermal power plants.

Distributed Energy Resources (DER) are revolutionizing how utilities plan and operate the grid, which ultimately influences the customer experience. Partially driven by increasing federal and state incentives for DERs, there is an increased interest to integrate solar generation facilities with the grid. To put this into perspective, interconnection applications in ComEd's territory increased more than 1,500% in the last three years and the accelerating trend continues. Most interconnection applications are for solar facilities that are less than 25kW in capacity (“Level 1” applications). The potential adverse impacts of each interconnection needs to be evaluated. Traditionally, a Level 1 technical screening is a manual process that involves reviewing the application against predetermined criteria set forth in the Illinois Administrative Code. ComEd has developed and deployed innovative solutions to automate such reviews by integrating data from multiple legacy ComEd systems and analyzing Level 1 applications to verify, that the applications meet the criteria to pass the technical screens. The automated technical screening process for most Level 1 application has facilitated improved management of complex utility scale commercial and industrial interconnection applications. This automation initiative has also enhanced customer service by significantly decreasing interconnection application turnaround time and ensuring a premier customer experience.

Blockchain technology with its secure mechanism of maintaining data and transactions in shared, immutable, distributed ledgers has become very relevant today and is increasingly used for financial applications. This paper proposes the use of consortium blockchain and smart contracts for secure, transparent and automated processing of student applications received by educational institutions. The students applying for admissions in educational institutions need assurance of a safe, secure and transparent platform that does not compromise their privacy. On the other hand, educational institutions too need assurance about the authenticity of the documents and the applicant. The use of consortium blockchain and smart contracts incorporating business logic for validating, verifying and filtering of valid applications provides a safe and secure platform for processing student applications. This paper looks at blockchain application beyond finance and explains how the student registration and admission process can be made safe and secure for all stakeholders. It promotes a seamless mechanism with reduced turnaround time and increased security and transparency.

As an extension of the cloud, a fog computing environment facilitates the deployment of Internet of Things (IoT) applications by shifting the computing, storage and networking services closer to the IoT devices, thus satisfying the delay and response time requirements. This paper aims to improve the overall task execution efficiency of IoT applications by appropriately selecting customized real-time tasks for execution at the fog layer. Specifically, we propose a fuzzy logic based task scheduling algorithm to divide the tasks between the fog and cloud layers in a fog-cloud computing framework. The algorithm selects appropriate processing units to execute the submitted tasks in the fog layer with heterogeneous resources, by exploiting the task requirements (e.g., computation, storage, bandwidth) and their constraints (e.g., deadline, data size). Simulation experiments demonstrate the efficacy of the proposed algorithm and its superior performance as compared to other existing algorithms in terms of success ratio of the tasks, makespan, average turnaround time, and delay rate.

In the field of cloud computing, efficient task scheduling plays a vital role. Cloud computing is an on-demand service where resources are shared virtually by the users to complete their tasks. The main concept of task scheduling is to execute the tasks in such an order that can enhance the performance with full resource utilization and minimum loss. Already a lot of scheduling algorithm exists targeting to various goals like reducing execution time, cost, makespan and increase resource utilization, load balancing, etc. In this work, we have mainly carried out the performance analysis and comparison among various well-known task scheduling algorithms. The algorithms are examined and implemented in a well-known cloud simulator (CloudSim) to observe their performances by varying different parameters. Finally, the performance metrics like makespan, waiting time, turnaround time, throughput, and load balancing are observed by plotting the results graphically. Along with the comparative analysis, this work also assists one to identify a better scheduling algorithm based on different scheduling constraints and criteria.

Accurate and sensitive taxonomic profiling is essential for any metagenomic analysis to reveal microbial community structure and for potential functional prediction. Antimicrobial resistance (AMR) detection is also a critical task in the clinical diagnosis of infection and antimicrobial therapy. By incorporating Oxford Nanopore Technologies (ONT) sequencing, users benefit from the high-confidence alignment of long reads for taxonomic classification, even among bacteria with similar genomes. Portable ONT devices, such as VolTRAX with MinION, allow short turnaround time for detection and can be used in a lightweight laboratory setting. However, error-prone ONT sequencing reads are still challenging for existing software for accurate taxonomic classification of microbes and detection of AMR down to the drug level. In this paper, we present MegaPath-Nano, the successor to NGS-based MegaPath. It is a high-precision compositional analysis software with drug-level AMR detection for ONT metagenomic sequencing data. MegaPath-Nano performs 1) thorough multi-level filtering against decoy and human reads while removing noisy alignments, 2) alignment-based taxonomic classification with RefSeq down to strain-level, with an alignment-reassignment algorithm to tackle the challenge of non-unique alignments, based on global alignment distribution, and 3) comprehensive downstream drug-level AMR detection, integrating five AMR databases. In our benchmarks using the Zymo metagenomic dataset, MegaPath-Nano performed better than other existing software for taxonomic classification. We also sequenced five real patient isolates using MinION to benchmark its performance of AMR detection. MegaPath-Nano was the most accurate and provided the most comprehensive output at both the drug and class level of AMR prediction against other state-of-the-art software. MegaPath-Nano is open-source and available at https://github.com/HKU-BAL/MegaPath-Nano.

Drug discovery is a costly process of searching for new candidate medications. Among its various stages, lead optimization easily consumes more than half of the pre-clinical budget. We propose an automated lead optimization workflow that uses data mining methods in components such as execution of molecular simulations, feature extraction, and clustering with convolutional variational autoencoder. The end-to-end execution produces protein-ligand binding affinity of atoms in the lead molecule which serves as metrics for identifying modifiable atoms. In contrast to known methods, our method provides new hints for drug modification hotspots which can be used to improve drug efficacy. Our workflow can potentially reduce the lead optimization turnaround time from months/years to several days compared with the conventional labor-intensive process and thus will become a valuable tool for medical researchers.

Ubiquitous computation, which promoted by Rapid development of Wireless Sensor Network (WSN) technologies cuts across many areas of modern day living, Internet of Things has been identified as one of Network Infrastructure in the next generation of application domains. An architecture based on cloud computing at the center, which contribute to highly flexibility and scalablity, is an extensively used scheme to construct IOT applications. With growing number of intelligent terminals and third part application accessing on the platform, the Qos problem caused by the large-scale concurrent access rise to the surface. To address this question, a self-adaption Multi-level Feedback Queue Scheduling policy, used to reduce mean turnaround time and complexity of scheduling, based on Recurrent Neural Network (RNN) is presented in this paper. Feature parameters of queues and tasks are used as input of network, the calculated parameters are exported to optimize queue parameters continuously. This research implement a prototype of this scheme. According To demonstrate the efficiency, this thesis give performance results from our prototype and other scheduling policy.

In this study, multi-layer scheduling optimization algorithms are proposed and validated based on historical vessel operation data in maritime terminals. The most relevant KPIs are average wait time, average turnaround time and berth occupancy rate presented in this study. Through the proposed optimization algorithms, the results shown that average wait time and turnaround time are significantly reduced with increasing of randomness threshold, which is a threshold to allow reschedules to buffer terminals. The average wait time and turnaround time are shortened by around 27.30 hrs (by 39.06%) and 39.41 hrs (by 27.18%), respectively. The berth occupancy rate of less utilized buffer terminals is also improved from 21.39% to 38.35%.

Cloud computing is a distributed computing model; which refers to manipulating, configuring, and accessing the applications online. It offers online data storage, infrastructure and application. Task scheduling is one of the most important issues in cloud environment, such as the tasks must be affected to the appropriate virtual machines considering different factors at the same time, which leads to considering the task scheduling problem an NP-complete problem. This article presents an improved ROUND ROBIN (RR) CPU scheduling algorithm with varying time quantum. The appropriate algorithm proven better than traditional RR and IRRVQ algorithms. The results show that the waiting time and turnaround time have been reduced in the proposed approach compared to traditional RR and IRRVQ

Recently, the Cloud Computing paradigm is becoming increasingly popular in supporting large-scale and complex workflow applications. The workflow scheduling problem, which refers to finding the most suitable resource for each task of the workflow to meet user defined quality of service (QoS), attracts considerable research attention. Multi-objective optimization algorithms in workflow scheduling have many limitations, e.g., the encoding schemes in most existing heuristic-based scheduling algorithms require prior experts' knowledge and thus they can be ineffective when scheduling workflows upon dynamic cloud infrastructures with real-time. To address this problem, we propose a novel Reinforcement-Learning-Based algorithm to multi-workflow scheduling over IaaS clouds. The proposed algorithm aims at optimizing make-span and dwell time and is to achieve a unique set of correlated equilibrium solution. In the experiment, our algorithm is evaluated for famous scientific workflow templates and real-world industrial IaaS cloud platforms by a simulation process and we compare our algorithm to the current state-of-the-art heuristic algorithms, e.g., NSGA-II, MOPSO, GTBGA. The result shows that our algorithm performs better than compared algorithm.

Panel-level packaging such as fan-out wafer-level packaging (FOWLP) has been a promising technology for a number of years now, primarily as a means of packaging semiconductor devices containing interconnect densities that exceed the capabilities of standard wafer-level chipscale packaging (WLCSP). One of the historical barriers to the broad adoption of panel-level packaging is the yield loss associated with “die drift”-die that shift from their designed nominal positions within each package during the manufacturing process. To break through this barrier, we introduce a novel die drift patterning technology that recognizes and adjusts for die drift, making “design during manufacturing” feasible and practical. However, both panel-level packaging and the die drift patterning methodology introduce physical verification challenges that are unfamiliar to most package designers. Panel-level packaging uses a GDSII or OASIS format for the package design, similar to integrated circuit (IC) design databases. Although design rule checking (DRC) is normally run on each individual unit GDSII file, the Die drift patterning process must also be simulated on a complete panel as one overall GDSII mask. This panel GDSII mask includes unique characteristics, with typically thousands of units requiring concurrent verification. The process is substantially more challenging than a classic unit design, where many repetitive GDSII cells exist within a hierarchy that can be used by the verification tools to improve runtimes. Deca collaborated with Mentor, a Siemens business (Mentor) to optimize physical verification for this panel GDSII mask verification. Together, they worked to identify operational impediments and implement optimizations to the verification toolsuite that enabled the platform to support verification of the die drift patterning technology for M-Series fan-out panel level packaging, while also achieving a reasonable turnaround time (TAT) for panel verification. This optimization utilizes both CPU scaling capabilities and a novel computational approach that accounts for the unique characteristics of a die drift patterning panel-level GDSII mask.

In-Flight Entertainment (IFE) datalink is one of the airport connectivity areas, where efforts are being made by different stakeholders to improve and update the entertainment services offered to the passengers. An important objective of IFE datalink is to increase the flight operation efficiency by managing IFE data transfer within turnaround time which is about 45 minutes. With the rapid advancements and innovation in multimedia applications and services, the IFE data size in the future will turn into terabytes, therefore, this transfer requires a multi-Gbps datalink in order to be completed within expected turnaround time. This paper focuses on simulation of IFE datalink communication scenario at an airport surface using Wi-Fi (802.11ac) technology, provides rules and guidelines on suitability of wireless datalinks for IFE update at the airports, studies QoS requirements, and performs optimization. Several aspects of the IFE datalink selection and deployment have been considered, such as airport operational areas, airport layouts, radio frequency, and data congestion before conducting a capacity and coverage analysis.

Point-of-Care Test (POCT) is considered as the diagnostic test which is administrated in the outside of the central laboratory ator near the patient location. Therefore, it can be stated that the use of POCT devices in the department of emergency represents a significant benefit to the management of the patient. It can be able to give proper governance to credible policy, safe practice, and effective process performance by the healthcare clinics. The major contribution of the POCT is it can give a huge improvement over the test result in the context of turnaround time. Here this research paper considers the process of development framework of POCT in the emergency department (ED). Moreover, the challenges faced by the clinical staff during the usage of previous POCT and the implementation process are also discussed to give a clear clarification about the essentiality of the developed framework of POCT.

Second-generation rapid diagnostic tests, e.g. Lab-on-a-Chip (LoC) molecular diagnostics can help decentralize medical diagnostics for infectious diseases and provide far wider coverage than existing methods. Furthermore, they can provide results far more quickly, enabling early diagnosis and prevention of epidemics. However, to ensure a short turnaround time in testing, a compact, ideally low-cost, automated readout method is required. Preferably, the readout should also have integrated means of communication for automatic report of the results. We demonstrate a readout solution based on an ESP12F microcontroller board to read results from a LoC device demonstrated in our previous works. The readout solution relies on an LED-photodiode setup to read the results of immunochromatographic test strips. We evaluated the device by comparing readouts of positive and negative strips. During the evaluation, the device prototype was run from batteries and transmitted results via Wi-Fi to a PC, where a MATLAB application processed, displayed and saved results.

Serverless computing emerges as a new paradigm to build cloud applications, in which developers write small functions that react to cloud infrastructure events, and cloud providers maintain all resources and schedule the functions in containers. Serverless computing thus enables developers to focus on their core business logic and leave server management and scaling to cloud providers. Unfortunately, existing serverless computing systems suffer from a key limitation that deprives them of enjoying significant speedups. Specifically, they treat each cloud function as a black box and are blind to which data the function reads or writes, therefore missing potentially huge optimization opportunities, such as caching data and colocating functions. We present Lambdata, a novel serverless computing system that enables developers to declare a cloud function's data intents, including both data read and data written. Once data intents are made explicit, Lambdata performs a variety of optimizations to improve speed, including caching data locally and scheduling functions based on code and data locality. Our evaluation of Lambdata shows that it achieves an average speedup of 1.51x on the turnaround time of practical workloads and reduces monetary cost by 16.5%.

To improve gross margins, the semiconductor industry is focused on the PPA (power, performance, area) matrix of the SOC. The current trend is to put more IPs on the chips to enable multiple functionalities to support various applications. To optimize PPA of such SOCs, multi voltage and multi power domain design techniques are used due to which the timing signoff of the chip has to be done on multiple corners and multiple modes (MCMM). Single voltage timing analysis is easier. With the multi-level supply voltage and dynamic scaling features, the timing analysis complexity increases because timing signoff has to be done on different voltages and cross-voltage paths. Multi-voltage designs need exhaustive analysis of cross voltage domain paths to make sure all worst-case paths are identified under all voltage combinations. With numerous operating PVT corners, timing analysis across corners is very challenging. Simultaneous multi-voltage aware analysis (SMVA) do the analysis of all cross-domain paths under all voltage scenarios in a single run, without the need for margining that can add pessimism. In this paper, we propose a machine learning model, based on bigrams of path stages, to predict the timing slack divergence and cell delays across voltages. We identified the circuit parameters which affects the cell delays due to voltage changes and thereby causing the differences in the endpoint arrival times. We use the timing analysis data of a given testcase at multiple voltages and with the use of Classification and regression tree (CART) approach we develop a predictive model for the new arrival times due to the change in voltages. Experimental results show that our model is able to predict the timing path slack divergence with ~97% accuracy at different voltages on both clock and data paths with a lower turnaround time including the cross-voltage timing paths.

Designing memories which embeds multiple features involves greater challenges. Implementing additional functions to memory degrades memory performance. Applications demanding more features along with high-performance create design challenges for memories. Designing feature rich memories using complete custom approach increases design turnaround time. Generally, memory implementation in Field Programable Gate Array (FPGA) follow hybrid approach of custom and digital design techniques. Defining optimal clock-distribution network in such feature rich designs is one of the most important aspect of high-speed Silicon-on-chip (SoC) design. This paper demonstrates the custom approach by prioritizing various features in terms of criticality of timing requirements and significantly improve the clock tree performance. This approach opens new possibility for designing feature rich memory with high chances of first pass design on silicon. These kinds of memories can be seen in the FPGA’s, CPU’s, GPU’s.

In this preliminary study, we consider the server allocation problem for edge computing system deployment. Our goal is to minimize the average turnaround time of application requests/tasks, generated by all mobile devices/users in a geographical region. We consider two approaches for edge cloud deployment: the flat deployment, where all edge clouds co-locate with the base stations, and the hierarchical deployment, where edge clouds can also co-locate with other system components besides the base stations. In the flat deployment, we demonstrate that the allocation of edge cloud servers should be balanced across all the base stations, if the application request arrival rates at the base stations are equal to each other. We also show that the hierarchical deployment approach has great potentials in minimizing the system's average turnaround time. We conduct various simulation studies using the CloudSim Plus platform to verify our theoretical results. The collective findings trough theoretical analysis and simulation results will provide useful guidance in practical edge computing system deployment.

As proposed accelerators target finer-grained chunks of computation and data movement, it becomes increasingly important to couple them tightly with the processor, avoiding long invocation delays. However, the large implementation design space of these Tightly-Coupled Accelerators (TCAs) makes it difficult to balance trade-offs between hardware complexity and accelerator performance. Previous performance models for accelerators focused on the penalties associated with loosely-coupled accelerators, which abstracted away many of the fine-grained interactions with complex out-of-order structures and program behaviors that have large impacts on TCA performance. In this paper, we introduce an analytical model that studies TCA behavior when interacting with the core, in the context of both high and low memory bandwidth applications supporting various levels of speculative and out of order (OoO) execution. Our analytical model reduces the turnaround time in early design stages when estimating performance gains over detailed simulation with tolerable error. We also discuss potential design choices that can impede the benefits that come with TCAs, and illuminate differences with traditional accelerators.

Lean Six Sigma (LSS) is a Quality Improvement (QI) method that integrates the speed of lean principles and robustness of Six Sigma (SS). LSS methodology, though invented and employed in manufacturing sectors initially, is also accepted widely across service sectors. The objective of this work is to reduce the Turnaround Time (TAT) of the laboratory tests in the pediatric emergency department of a tertiary-care hospital in India. The applicability of LSS DMAIC (Define-Measure-Analyze-Improve-Control) methodology to reduce the TAT is explored. During the pre-intervention period, a total of 44 samples were sent to the lab for testing, and the average TAT was found to be 69 minutes. A multidisciplinary team of experts constructed the Value Stream Map (VSM), and each step in the process was analyzed. Through brainstorming and 5-why analysis, the root causes of delay were identified, and the corresponding improvement actions were finalized. During the course of the work, various lean wastes in the process were also identified and eliminated. A new process map was designed, and the improvised process was simulated using the simulation software ARENA and results were obtained. The results of the improvised process show a substantial reduction in TAT. The reduced TAT will decrease the overcrowding of patients and thus will result in smooth patient flow. Hence, through the systematic LSS approach, the opportunities for improvements can be identified and eliminated for better operational efficiency and patient satisfaction in the healthcare sector.