Exploring the interoperability of remote GPGPU virtualization using rCUDA and directive-based programming models

Castelló, Adrián; Peña, Antonio J.; Mayo, Rafael; Planas, Judit; Quintana–Ort́ı, Enrique S.; Balaji, Pavan

doi:10.1007/s11227-016-1791-y

Cited by 3 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remote API technique [11][12][13] redirects GPU call requests to a remote GPU-equipped node. This mechanism proves effective in a cluster with limited number of GPU-equipped nodes.…”

Section: Gpu Underutilization: This Case Can Be Divided Into Two Cate...mentioning

confidence: 99%

See 1 more Smart Citation

Adaptive and transparent task scheduling of GPU‐powered clusters

Yang

et al. 2020

Concurrency and Computation

View full text Add to dashboard Cite

GPGPU-powered supercomputers are vital for various science and engineering applications. On each cluster node, the GPU works as a coprocessor of the CPU, and the computing task runs alternatively on CPU and GPU. Due to this characteristic, traditional task scheduling strategy tends to result in significant workload imbalance and underutilization of GPUs. We design an adaptive scheduling strategy to alleviate such imbalance and underutilization. Our strategy proposes to logically treats all GPUs in the cluster as a whole. Every cluster node maintains a local information table of all GPUs. Once a GPU call request is received, a node selects a GPU to run the task in an adaptive manner based on this table. In addition, our strategy does not rely on a global queue, and thus avoids excessive internode communication overhead. Moreover, we encapsulate our strategy into an intermedia module between the cluster and users. Consequently, underlying details of task scheduling is transparent to users, which enhances usability. We validate our strategy through experiments.

show abstract

“…Remote API technique [11][12][13] redirects GPU call requests to a remote GPU-equipped node. This mechanism proves effective in a cluster with limited number of GPU-equipped nodes.…”

Section: Gpu Underutilization: This Case Can Be Divided Into Two Cate...mentioning

confidence: 99%

“…In response to the challenges of building an efficient GPU cluster, the enabling technology of GPU clusters has been an active research area. Remote API technique 11–13 redirects GPU call requests to a remote GPU‐equipped node. This mechanism proves effective in a cluster with limited number of GPU‐equipped nodes.…”

Section: Introductionmentioning

confidence: 99%

Adaptive and transparent task scheduling of GPU‐powered clusters

Yang

et al. 2020

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…In "Exploring the interoperability of remote GPGPU virtualization using rCUDA and directive-based programming models" [9], Castelló et al study the integration of two directive-based programming models, OmpSs and OpenACC, to make use of remote virtualized CUDA devices through rCUDA. This work moves towards a scenario where applications do not need a full manual rewrite for moving from local GPGPU to the use of remote GPGPUs.…”

Section: Special Issue Presentationmentioning

confidence: 99%