A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Filters
iGPU: Exception support and speculative execution on GPUs
2012
2012 39th Annual International Symposium on Computer Architecture (ISCA)
Preserved Fulltext
Exception support and speculative execution are the next steps to expand the scope and improve the usability of GPUs. ...
This paper builds on two related insights to provide a unified lightweight mechanism for supporting exceptions and speculation on GPUs. ...
Acknowledgments We thank the anonymous reviewers, the Vertical group, and Mark Gebhart for comments. Many thanks to Steve ...
doi:10.1109/isca.2012.6237007
dblp:conf/isca/MenonKS12
fatcat:l7js6vnw2bcctduetmpjacmrdy
iGPU
2012
SIGARCH Computer Architecture News
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Exception support and speculative execution are the next steps to expand the scope and improve the usability of GPUs. ...
This paper builds on two related insights to provide a unified lightweight mechanism for supporting exceptions and speculation on GPUs. ...
Acknowledgments We thank the anonymous reviewers, the Vertical group, and Mark Gebhart for comments. Many thanks to Steve ...
doi:10.1145/2366231.2337168
fatcat:tylnlqwudrgjjllwvx5oxhkqiq
iGPU-Accelerated Pattern Matching on Event Streams
2022
Data Management on New Hardware
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In addition, hardware accelerators based on dedicated GPUs also offer limited support due to the overhead of transferring data between their local and main memory. ...
This paper presents the first full-fledged implementation of pattern matching cooperatively using iGPUs and CPUs. ...
ACKNOWLEDGMENTS This work has been supported by the German Research Foundation (DFG) under grant no. SE 553/9-1. ...
doi:10.1145/3533737.3535099
fatcat:tcrj6saitjh7jbmte3o6dd5soe
Pattern of Pediatric Toxicity in Saudi Arabia-Eastern Province (Incidence, Demographics and Predisposing Factors)
2015
Pediatrics & Therapeutics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Exception support and speculative execution are the next steps to expand the scope and improve the usability of GPUs. ...
This paper builds on two related insights to provide a unified lightweight mechanism for supporting exceptions and speculation on GPUs. ...
Acknowledgments We thank the anonymous reviewers, the Vertical group, and Mark Gebhart for comments. Many thanks to Steve ...
doi:10.4172/2161-0665.1000220
fatcat:5sdip55i6zepzj4yd2gtzvupri
APUNet: Revitalizing GPU as Packet Processing Accelerator
2017
Symposium on Networked Systems Design and Implementation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
to the Proceedings of the 14th USENIX Symposium on Networked Systems Design and Implementation is sponsored by USENIX. ...
Acknowledgments We would like to thank our shepherd Fabián Bustamante and anonymous reviewers of NSDI'17 for their insightful comments and suggestions on the paper. ...
We also thank Anuj Kalia at CMU for in-depth discussion on experiment results and further suggestions, and Seong Hwan Kim and Dipak Bhattacharyya at AMD, Inc., for their help in understanding the internals ...
dblp:conf/nsdi/GoJMHP17
fatcat:3ght6653hfhbfmawnmz2likfkm
Enabling GPGPU Low-Level Hardware Explorations with MIAOW
2015
ACM Transactions on Architecture and Code Optimization (TACO)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
Further, as GPGPU research evolves, design ideas and modifications demand detailed estimates of impact on overall area and power. ...
Today's tools for GPU analysis include simulators like GPGPU-Sim, Multi2Sim, and Barra. ...
We have also implemented in MIAOW the modifications for idempotent re-execution [Menon et al. 2012 ] from the iGPU design, which supports timing speculation. ...
doi:10.1145/2764908
fatcat:utj6prgm2zcctlb36ikgejny2e
Microarchitectural Attacks in Heterogeneous Systems: A Survey
2022
ACM Computing Surveys
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this survey article, we consider the security of heterogeneous systems against microarchitectural attacks, with a focus on covert- and side-channel attacks, as well as fault injection attacks. ...
We review works that have explored the vulnerability of the individual accelerators (such as Graphical Processing Units, GPUs and Field Programmable Gate Arrays, FPGAs) against these attacks, as well as ...
The work is supported by the National Science Foundation under CNS-2053383, CNS-2130978, and CNS-1955650 grants. ...
doi:10.1145/3544102
fatcat:dm3l4etg5vealdqyodhuopjhgy
Chimera
2015
Proceedings of the Twentieth International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS '15
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Preemptive multitasking on CPUs has been primarily supported through context switching. However, the same preemption strategy incurs substantial overhead due to the large context in GPUs. ...
Chimera first introduces streaming multiprocessor (SM) flushing, which can instantly preempt an SM by detecting and exploiting idempotent execution. ...
Khudia, Shruti Padmanabha, and Ankit Sethia for their valuable comments and feedbacks. This work is supported by the National Science Foundation under grants CNS-0964478 and CCF-1438996. ...
doi:10.1145/2694344.2694346
dblp:conf/asplos/ParkPM15
fatcat:kaqig6ktkjhfrjxgbaavpiy33i
GDM
2014
Performance Evaluation Review
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Device memory is the primary onboard DRAM storage for the computation performed on GPU. ...
Managing device memory space at application level becomes even more difficult when there are multiple applica-1 We use GPU and GPGPU interchangeably, with the latter emphasizing more on general-purpose ...
Inc. and the anonymous reviewers for their help and feedback. This work was partially supported by the National Science Foundation under grants CCF-0913050, OCI-1147522, and CNS-1162165. ...
doi:10.1145/2637364.2592002
fatcat:zcpzph5qfzh3jczn7nsnj2iruy
Fine-grained containment domains for throughput processors
[article]
2017
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
A similar approach has been proposed for GPUs to support exceptions as well as speculative execution [57] . ...
Recent proposals on in-order processors [56] and GPUs [57] revisit the concept of idempotence to implement software-based fault recovery on processors without hardware speculation support. ...
Chapter 5 Methodology In this chapter, we discuss the methodologies we use to evaluate the applicability of fine-grained containment domains to timing speculation on GPUs. ...
doi:10.15781/t2h41js69
fatcat:5usb4eurr5bcpnlzbnfynt2vnu
Accelerating Event Stream Processing in On- and Offline Systems
2022
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
While this naturally ensures unified query semantics for on- and offline processing, the costs for reading the entire stream from non-volatile storage quickly dominate the overall processing costs. ...
Both modes are supported by widely used scale-out general-purpose stream processing engines (SPEs) like Apache Flink or Spark Streaming. ...
However, without the help and support from my colleagues, family, and friends, this work would not have been possible.
Acknowledgments viii ...
doi:10.17192/z2022.0078
fatcat:x64m6dluzjhhzi7mycwod7lkhq
Opportunistic acceleration of array-centric Python computation in heterogeneous environments
2020
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
However, such managed program execution comes with a significant performance penalty arising from the interpreter having to decode and dispatch based on dynamic type checking. ...
Parallel GPU kernels are customised to the runtime dependence graph, JIT compiled and executed. A systematic analysis of the execution speed of loop nests is performed using [...] ...
Samadi et al [126] and Wang et al [156] use speculative kernel execution on GPUs and require post-execution correctness checking before committing the results. ...
doi:10.5525/gla.thesis.82011
fatcat:cokrisg5fncu7a2z4iubykf6lm
Software-hardware co-design for energy efficient datacenter computing
2019
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
GNoM facilitates the design of MemcachedGPU, an end-to-end Memcached implementation on contemporary Ethernet and GPU hardware. ...
GNoM highlights limitations in the traditional GPU programming model, which relies on a CPU for managing GPU tasks. Consequently, the CPU may be unnecessarily invol [...] ...
[117] propose iGPU, a set of compiler, ISA, and hardware extensions that enable preemption support and speculative execution on GPUs. iGPU identifies and exploits sparse idempotent regions to minimize ...
doi:10.14288/1.0384819
fatcat:x2f5bfzndjhgdpcxlonwsj7uoe
Ανάπτυξη Συστοιχιών Υπολογιστών (Beowulf Clusters) με τη Χρήση Desktop Η/Υ, Graphics Processing Units (GPUs) και Ελεύθερου Λογισμικού για Παράλληλη Επεξεργασία (Parallel Computing) με Εφαρμογές στη Βιοϊατρική Τεχνολογία
[article]
2021
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Intel Wide Dynamic Execution
Πρόκειται για έναν συνδυασμό τεχνικών (data flow analysis, speculative execution, out-of-
order execution και superscalar (=ανάλυση ροής δεδομένων, υποθετική εκτέλεση,
εκτέλεση ...
(IGPUs): χρησιμοποιούν ένα ποσοστό της RAM του συστήματος. ...
doi:10.26240/heal.ntua.21315
fatcat:7zl44e6xdbgs3idml7b3nj7u4q