2019
DOI: 10.1145/3419973
|View full text |Cite
|
Sign up to set email alerts
|

Highly Concurrent Latency-tolerant Register Files for GPUs

Abstract: Graphics Processing Units (GPUs) employ large register files to accommodate all active threads and accelerate context switching. Unfortunately, register files are a scalability bottleneck for future GPUs due to long access latency, high power consumption, and large silicon area provisioning. Prior work proposes hierarchical register file to reduce the register file power consumption by caching registers in a smaller register file cache. Unfortunately, this approach does not improve register access latency due … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3

Citation Types

0
3
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 8 publications
(3 citation statements)
references
References 80 publications
0
3
0
Order By: Relevance
“…Additionally, AVA extend the use of the associated counters to decide option to perform swaps between Physical and VVRs. The idea of using memory to provide a backing store to the register file has been has also been widely studied for outof-order cores [30] , VLIW processors [41], and GPUs [23], [34]. In this work, we apply it to VPs as a key mechanism to offer a variety of MVL configurations.…”
Section: Related Workmentioning
confidence: 99%
“…Additionally, AVA extend the use of the associated counters to decide option to perform swaps between Physical and VVRs. The idea of using memory to provide a backing store to the register file has been has also been widely studied for outof-order cores [30] , VLIW processors [41], and GPUs [23], [34]. In this work, we apply it to VPs as a key mechanism to offer a variety of MVL configurations.…”
Section: Related Workmentioning
confidence: 99%
“…Additionally, AVA extend the use of the associated counters to decide the best option to perform swaps between Physical and VVRs. The idea of using memory to provide a backing store to the register file has been has also been widely studied for outof-order cores [30] , VLIW processors [41], and GPUs [23], [34]. In this work, we apply it to VPs as a key mechanism to offer a variety of MVL configurations.…”
Section: Related Workmentioning
confidence: 99%
“…The idea of using memory to provide a backing store to the register file has been has also been widely studied for out-of-order cores [70], VLIW processors [71], and GPUs [72] [73]. In this work, we apply it to vector processors as a key mechanism to offer a variety of MVL configurations.…”
Section: Related Workmentioning
confidence: 99%