2017
DOI: 10.1109/access.2017.2661582
|View full text |Cite
|
Sign up to set email alerts
|

Virtualized Execution Runtime for FPGA Accelerators in the Cloud

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
27
0
1

Year Published

2018
2018
2023
2023

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 61 publications
(28 citation statements)
references
References 21 publications
0
27
0
1
Order By: Relevance
“…Only very recent research is investigating FPGA virtualization in the spatial domain. Asiatici et al [13] proposed dynamic scheduling which can take advantage of the free slots available at run-time to improve utilization and thus the performance. This scheduling approach maximizes the number of pipeline instances when a task is created and keeps this allocation untouched until task completion.…”
Section: Schedulingmentioning
confidence: 99%
See 1 more Smart Citation
“…Only very recent research is investigating FPGA virtualization in the spatial domain. Asiatici et al [13] proposed dynamic scheduling which can take advantage of the free slots available at run-time to improve utilization and thus the performance. This scheduling approach maximizes the number of pipeline instances when a task is created and keeps this allocation untouched until task completion.…”
Section: Schedulingmentioning
confidence: 99%
“…Thus, infrastructure and resource management techniques are considered for this level. Examples include VMM support [11], [12], run-time systems [13] and Shells (also called FPGA OS and Hypervisor-vFPGA approach) which are used to serve multiple concurrent user accelerators [2], [11], [14]- [16].…”
Section: Introductionmentioning
confidence: 99%
“…But the required support for the network layer consume a reasonable resource of FPGA. However, Asiatici et al [89] developed a lightweight version featuring high-end application program interface (API), with a simpler execution model and shared memory. They proved their concept by measuring the marginal performance overhead.…”
Section: Shellsmentioning
confidence: 99%
“…Largely, scheduling techniques fall in non-preemptive category, which is fundamentally a time domain optimization. However, a dynamic approach has recently been introduced in [89], which takes advantage of the empty slots and keeps the utilization balanced on the run time. This dynamic scheduling technique enables the multi-tenancy like none other, as it gives the power of increasing or decreasing the resources usage, as per the workload requirement.…”
Section: Schedulingmentioning
confidence: 99%
See 1 more Smart Citation