Abstract With the evolution of high‐performance computing toward heterogeneous, massively parallel systems, applications have developed new checkpoint and restart necessities. Whether due to a failure in execution or migration application processes different machines, checkpointing tools must be able operate heterogeneous environments. However, some data manipulated by are not truly portable. Examples these include opaque state (e.g. structures for communications support) diversity...

Serverless computing is an emerging paradigm that has gained a lot of relevance in recent years, as it allows users to consume resources without worrying about the underlying infrastructure and pay only for what they actually use. Most current services implement this typically rely on Function-as-a-Service (FaaS) model, which works perfectly simple applications based stateless functions triggered by specific events. However, these are not designed run more complex with intricate...

Summary: MSAProbs is a state-of-the-art protein multiple sequence alignment tool based on hidden Markov models. It can achieve high accuracy at the expense of relatively long runtimes for large-scale input datasets. In this work we present MSAProbs-MPI, distributed-memory parallel version multithreaded that able to reduce by exploiting compute capabilities common multicore CPU clusters. Our performance evaluation cluster with 32 nodes (each containing two Intel Haswell processors) shows...

Land consolidation is a strategic instrument for rural planning and thus economic development in the Spanish region of Galicia. This paper describes an experimental system embedded GIS environment to aid engineers develop land plans. The supports all stages plan many functionalities are implemented as heuristic processes based on expert knowledge advice. overall aim overcome administrative technical problems traditional procedures. provides integrated framework management spatial...

The rising interest in Java for High Performance Computing (HPC) is based on the appealing features of this language programming multi-core cluster architectures, particularly built-in networking and multithreading support, continuous increase Virtual Machine (JVM) performance. However, its adoption area being delayed by lack analysis existing options HPC evaluations their performance, as well unawareness current research projects field, whose solutions are needed order to boost embracement...

SUMMARY Cloud computing is offering new approaches for High Performance Computing (HPC) as it provides dynamically scalable resources a service over the Internet. In addition, General‐Purpose computation on Graphical Processing Units (GPGPU) has gained much attention from scientific in multiple domains, thus becoming an important programming model HPC. Compute Unified Device Architecture (CUDA) been established popular GPGPUs, removing need using graphics APIs applications. Open Language...

The recognition of program constructs that are frequently used by software developers is a powerful mechanism for optimizing and parallelizing compilers to improve the performance object code. development techniques automatic computational kernels such as inductions, reductions array recurrences has been an intensive research area in scope compiler technology during 90's. This article presents new framework that, unlike previous focus on specific isolated kernels, recognizes comprehensive...

To efficiently scale dense linear algebra problems to future exascale systems, communication cost must be avoided or overlapped. Communication-avoiding 2.5D algorithms improve scalability by reducing inter-processor data transfer volume at the of extra memory usage. Communication overlap attempts hide messaging latency pipelining messages and overlapping with computational work. We study interaction compatibility these two techniques for matrix multiplication (Cannon SUMMA), triangular...

On-chip power consumption is one of the fundamental challenges current technology scaling. Cache memories consume a sizable part this power, particularly due to leakage energy. STT-RAM several new memory technologies that have been proposed in order improve while preserving performance. It features high density and low leakage, but at expense write energy This article explores use STT-RAM--based scratchpad trade nonvolatility exchange for faster less energetically expensive accesses, making...

Development of new methods to detect pairwise epistasis, such as SNP-SNP interactions, in Genome-Wide Association Studies is an important task bioinformatics they can help explain genetic influences on diseases. As these studies are time consuming operations, some tools exploit the characteristics different hardware accelerators (such GPUs and Xeon Phi coprocessors) reduce runtime. Nevertheless, all approaches not able efficiently whole computational capacity modern clusters that contain...

Java is a valuable and emerging alternative for the development of parallel applications, thanks to availability several message-passing libraries its full multithreading support. The combination both shared distributed memory programming an interesting option multi-core systems. However, concerns about performance are hindering adoption in this field, although it difficult evaluate accurately due lack standard benchmarks Java. This paper presents NPB-MPJ, first extensive implementation NAS...

The growing complexity in computer system hierarchies due to the increase number of cores per processor, levels cache (some them shared) and processors node, as well high-speed interconnects, demands use new optimization techniques libraries that take advantage their features. In this paper Servet, a suite benchmarks focused on detecting set parameters with high influence overall performance multicore systems, is presented. These are able detect hierarchy, including size which caches shared...

This article presents MarDRe, a de novo cloud-ready duplicate and near-duplicate removal tool that can process single- paired-end reads from FASTQ/FASTA datasets. MarDRe takes advantage of the widely adopted MapReduce programming model to fully exploit Big Data technologies on cloud-based infrastructures. Written in Java maximize cross-platform compatibility, is built upon open-source Apache Hadoop project, most popular distributed computing framework for scalable processing. On 16-node...