Performance Evaluation of Hypervisors for HPC Applications

Beserra, David; Oliveira, Felipe; Araújo, Jean; Fernandes, Felipe; Araújo, Alberto Filipe; Endo, Patrícia Takako; Maciel, Paulo; Moreno, Edward David

doi:10.1109/smc.2015.156

Cited by 11 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Their evaluations indicate that the KVM HV with full virtualization VM increases the average processing time substantially (on the order of 50% increase) compared to container and native processing; whereas, the KVM HV with para-virtualization achieves nearly the same low average processing times as the container based processing. Similarly, the older study by Beserra et al [147] has found benefits of para-virtualization for HPC applications.…”

Section: C: Para-virtualizationmentioning

confidence: 77%

Operating Systems and Hypervisors for Network Functions: A Survey of Enabling Technologies and Research Studies

et al. 2022

View full text Add to dashboard Cite

Scalable and flexible communication networks increasingly conduct the packet processing for Network Functions (NFs) in General Purpose Computing (GPC) platforms. The input/output (I/O)-intensive and latency-sensitive packet processing is challenging for the operating systems and hypervisors running on GPC platforms. This article surveys the existing enabling technologies and research studies on operating system and hypervisor aspects that directly influence the packet processing for NFs on GPC platforms. We organize this survey according to the main categories abstraction approach, memory access, and I/O strategy. We further categorize abstraction approach technologies and research studies into the categories operation systems, hypervisors, and containers. We partition the memory access category into the two subcategories of memory allocation and memory access, while we partition the I/O strategy category into the sub-categories I/O device virtualization and I/O device access. Our survey gives a comprehensive summary of the capabilities and limitations of the existing enabling technologies and researched approaches for abstraction, memory access, and I/O for NF packet processing. We outline critical future research directions for advancing NF packet processing on GPC platforms.

show abstract

Section: C: Para-virtualizationmentioning

confidence: 77%

Operating Systems and Hypervisors for Network Functions: A Survey of Enabling Technologies and Research Studies

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The estimation of the computing capacity of non-HPC clusters to tackle HPC applications is a common and ongoing research problem. Related works estimate the capacity of virtual clusters built on top of containers or virtual machines over cloud, [5][6][7][8][9] Internet of things (IoT), 10,11 workstations, 12 and desktops 13 infrastructures. Bare-metal cluster deployments on top of workstations 14 and desktops 15 have also been considered.…”

Section: Related Workmentioning

confidence: 99%

Estimating the overhead and coupling of scientific computing clusters

Vivas

Castro

2021

SIMULATION

View full text Add to dashboard Cite

Since simulation became the third pillar of scientific research, several forms of computers have become available to drive computer aided simulations, and nowadays, clusters are the most popular type of computers supporting these tasks. For instance, cluster settings, such as the so-called supercomputers, cluster of workstations (COW), cluster of desktops (COD), and cluster of virtual machines (COV) have been considered in literature to embrace a variety of scientific applications. However, those scientific applications categorized as high-performance computing (HPC) are conceptually restricted to be addressed only by supercomputers. In this aspect, we introduce the notions of cluster overhead and cluster coupling to assess the capacity of non-HPC systems to handle HPC applications. We also compare the cluster overhead with an existing measure of overhead in computing systems, the total parallel overhead, to explain the correctness of our methodology. The evaluation of capacity considers the seven dwarfs of scientific computing, which are well-known, scientific computing building blocks considered in the development of HPC applications. The evaluation of these building blocks provides insights regarding the strengths and weaknesses of non-HPC systems to deal with future HPC applications developed with one or a combination of these algorithmic building blocks.

show abstract

“…Nos testes com utilizac ¸ão de vários nós as diferenc ¸as se tornaram mais evidentes e o Xen ficou com a pior performance devido ao uso da rede. No artigo [Beserra 2016] o objetivo era descobrir qual hypervisor possui a melhor performance para fazer a virtualizac ¸ão com foco em computac ¸ão de alto desempenho e em todos os casos o KVM se mostrou melhor para se aplicar ao uso de computac ¸ão de alto desempenho.…”

Section: Trabalhos Relacionadosunclassified

Análise do Comportamento de Aplicações Paralelas em Ambientes de Computação de Alto Desempenho Virtualizados

Ferruzzi¹

2020

Anais Do XVIII Workshop Em Clouds E Aplicações (WCGA 2020)

View full text Add to dashboard Cite

A computação em nuvem tem se tornado cada vez mais frequente em empresas e universidades, principalmente devido ao fato da nuvem computacional ser mais facilmente administrada uma vez que seu ambiente esteja devidamente configurado. Entretanto, há resistência na implantação de ambientes de computação de alto desempenho em nuvem e isso, provavelmente, está associado ao fato da nuvem precisar de um hypervisor para a administração das máquinas virtuais, resultando em uma perda de desempenho causado pela camada de virtualização. Por outro lado, as tecnologias de virtualização estão em constante avanço, tornando essa perda de desempenho menor com o passar dos anos. Sendo assim, o trabalho faz a análise do desempenho de aplicações paralelas e da infraestrutura utilizada além de avaliar as características dos hypervisors XenServer e KVM com o propósito de comparar o desempenho de cada ambiente e identificar qual a melhor hypervisor a ser utilizada na criação de um ambiente de computação de alto desempenho virtualizado.

show abstract

Performance Evaluation of Hypervisors for HPC Applications

Cited by 11 publications

References 7 publications

Operating Systems and Hypervisors for Network Functions: A Survey of Enabling Technologies and Research Studies

Operating Systems and Hypervisors for Network Functions: A Survey of Enabling Technologies and Research Studies

Estimating the overhead and coupling of scientific computing clusters

Análise do Comportamento de Aplicações Paralelas em Ambientes de Computação de Alto Desempenho Virtualizados

Contact Info

Product

Resources

About