research-article

Collaborative Computing for Heterogeneous Integrated Systems

Authors:

Juan Gómez-Luna,

Wen-mei HwuAuthors Info & Claims

ICPE '17: Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering

Pages 385 - 388

https://doi.org/10.1145/3030207.3030244

Published: 17 April 2017 Publication History

Abstract

Computing systems today typically employ, in addition to powerful CPUs, various types of specialized devices such as Graphics Processing Units (GPUs) and Field-Programmable Gate Arrays (FPGAs). Such heterogeneous systems are evolving towards tighter integration of devices for improved performance and reduced energy consumption. Compared to traditional use of GPUs and FPGAs as offload accelerators, this tight integration enables close collaboration between processors which is important for better utilization of system resources and higher performance. Programming interfaces are also adapting rapidly to tightly integrated heterogeneous platforms by introducing features such as shared virtual memory, memory coherence, and system-wide atomics, making collaborative computing among different devices even more practical.

In this paper, we survey current integrated heterogeneous systems and corresponding collaboration techniques. We evaluate the impact of collaborative computing on two heterogeneous integrated systems, CPU-GPU and CPU-FPGA, using OpenCL. Finally, we discuss the limitation of OpenCL and envision what suitable programming languages for collaborative computing will look like.

References

[1]

Cache Coherent Interconnect for Accelerators (CCIX). http://www.ccixconsortium.com, 2016.

[2]

Altera. Altera's User-Customizable ARM-Based SoC, 2015.

[3]

C. Augonnet et al. StarPU: a unified platform for task scheduling on heterogeneous multicore architectures. CCPE, 23(2):187--198, 2011.

Digital Library

[4]

Bruce Wile. IBM Systems and Technology Group. Coherent Accelerator Processor Interface (CAPI) for POWER8 systems. White paper, September 2014.

[5]

A. M. Caulfield et al. A cloud-scale acceleration architecture. In ISCA, 2016.

[6]

L.-W. Chang et al. Efficient kernel synthesis for performance portable programming. In MICRO, 2016.

[7]

L.-W. Chang et al. A programming system for future proofing performance critical libraries. In PPoPP, 2016.

Digital Library

[8]

E. S. Chung et al. Single-chip heterogeneous computing: Does the future include custom logic, FPGAs, and GPGPUs? In MICRO, 2010.

Digital Library

[9]

A. Duran et al. OmpSs: a proposal for programming heterogeneous multi-core architectures. PPL, 21(02):173--193, 2011.

[10]

V. Garcia-Flores et al. Evaluating the effect of last-level cache sharing on integrated GPU-CPU systems with heterogeneous applications. In IISWC, 2016.

[11]

J. Gómez-Luna et al. Chai: Collaborative heterogeneous applications for integrated-\\architectures. In ISPASS, 2017 (in press).

[12]

W.-m. W. Hwu. Heterogeneous System Architecture: A New Compute Platform Infrastructure. Morgan Kaufman, 2015.

[13]

Intel. Intel FPGA SDK for OpenCL. Programming Guide, October 2016.

[14]

A. Morad et al. Generalized multiAmdahl: Optimization of heterogeneous multi-accelerator SoC. IEEE CAL, 13(1):37--40, Jan. 2014.

Digital Library

[15]

S. Mukherjee et al. Exploring the features of OpenCL 2.0. In IWOCL, 2015.

Digital Library

[16]

S. Mukherjee et al. A comprehensive performance analysis of HSA and OpenCL 2.0. In ISPASS, 2016.

[17]

NVIDIA. NVIDIA Tesla P100. White paper, 2016.

[18]

PK Gupta. Intel. Xeon

[19]

FPGA Platform for the Data Center, June 2015.

[20]

A. Putnam et al. A reconfigurable fabric for accelerating large-scale datacenter services. In ISCA, 2014.

[21]

J. Ragan-Kelley et al. Halide: A language and compiler for optimizing parallelism, locality, and recomputation in image processing pipelines. In PLDI, 2013.

Digital Library

[22]

Y. Sun et al. Hetero-Mark, a benchmark suite for CPU-GPU collaborative computing. In IISWC, 2016.

[23]

Xilinx. Zynq UltraScale

[24]

MPSoCs. White Paper, June 2016.

Cited By

Dakić VMršić LKunić ZĐambić G(2024)Evaluating ARM and RISC-V Architectures for High-Performance Computing with Docker and KubernetesElectronics10.3390/electronics1317349413:17(3494)Online publication date: 3-Sep-2024
https://doi.org/10.3390/electronics13173494
Zeng QDu YHuang KLeung K(2021)Energy-Efficient Resource Management for Federated Edge Learning With CPU-GPU Heterogeneous ComputingIEEE Transactions on Wireless Communications10.1109/TWC.2021.308891020:12(7947-7962)Online publication date: Dec-2021
https://doi.org/10.1109/TWC.2021.3088910
Freytag GSerpa MLima JRech PNavaux P(2021)Collaborative execution of fluid flow simulation using non-uniform decomposition on heterogeneous architecturesJournal of Parallel and Distributed Computing10.1016/j.jpdc.2021.02.006152(11-20)Online publication date: Jun-2021
https://doi.org/10.1016/j.jpdc.2021.02.006
Show More Cited By

Index Terms

Collaborative Computing for Heterogeneous Integrated Systems
1. General and reference
  1. Cross-computing tools and techniques
    1. Performance

Recommendations

A Halide-based Synergistic Computing Framework for Heterogeneous Systems

New programming models have been developed to embrace contemporary heterogeneous machines, each of which may contain several types of processors, e.g., CPUs, GPUs, FPGAs and ASICs. Unlike the conventional ones, which use separate programming schemes for ...
On the Efficacy of a Fused CPU+GPU Processor (or APU) for Parallel Computing
SAAHPC '11: Proceedings of the 2011 Symposium on Application Accelerators in High-Performance Computing

The graphics processing unit (GPU) has made significant strides as an accelerator in parallel computing. However, because the GPU has resided out on PCIe as a discrete device, the performance of GPU applications can be bottlenecked by data transfers ...
Adaptive heterogeneous scheduling for integrated GPUs
PACT '14: Proceedings of the 23rd international conference on Parallel architectures and compilation

Many processors today integrate a CPU and GPU on the same die, which allows them to share resources like physical memory and lowers the cost of CPU-GPU communication. As a consequence, programmers can effectively utilize both the CPU and GPU to execute ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

ICPE '17: Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering

April 2017

450 pages

ISBN:9781450344043

DOI:10.1145/3030207

General Chairs:
Walter Binder
Università della Svizzera Italiana, Switzerland
,
Vittorio Cortellessa
Università dell'Aquila, Italy
,
Program Chairs:
Anne Koziolek
Karlsruhe Institute of Technology, Germany
,
Evgenia Smirni
College of William & Mary, USA
,
Meikel Poess
Oracle Corporation, USA

Copyright © 2017 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

SIGMETRICS: ACM Special Interest Group on Measurement and Evaluation
SIGSOFT: ACM Special Interest Group on Software Engineering
SPEC: SPEC Research Group

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 17 April 2017

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Conference

ICPE '17

Sponsor:

ICPE '17: ACM/SPEC International Conference on Performance Engineering

April 22 - 26, 2017

L'Aquila, Italy

Acceptance Rates

ICPE '17 Paper Acceptance Rate 27 of 83 submissions, 33%;

Overall Acceptance Rate 252 of 851 submissions, 30%

Upcoming Conference

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

10
Total Citations
View Citations
431
Total Downloads

Downloads (Last 12 months)35
Downloads (Last 6 weeks)3

Reflects downloads up to 08 Mar 2025

Other Metrics

View Author Metrics

Citations

Cited By

Dakić VMršić LKunić ZĐambić G(2024)Evaluating ARM and RISC-V Architectures for High-Performance Computing with Docker and KubernetesElectronics10.3390/electronics1317349413:17(3494)Online publication date: 3-Sep-2024
https://doi.org/10.3390/electronics13173494
Zeng QDu YHuang KLeung K(2021)Energy-Efficient Resource Management for Federated Edge Learning With CPU-GPU Heterogeneous ComputingIEEE Transactions on Wireless Communications10.1109/TWC.2021.308891020:12(7947-7962)Online publication date: Dec-2021
https://doi.org/10.1109/TWC.2021.3088910
Freytag GSerpa MLima JRech PNavaux P(2021)Collaborative execution of fluid flow simulation using non-uniform decomposition on heterogeneous architecturesJournal of Parallel and Distributed Computing10.1016/j.jpdc.2021.02.006152(11-20)Online publication date: Jun-2021
https://doi.org/10.1016/j.jpdc.2021.02.006
Jiang JWang ZLiu XGómez-Luna JGuan NDeng QZhang WMutlu ONeuendorffer SShannon L(2020)BoyiProceedings of the 2020 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays10.1145/3373087.3375313(299-309)Online publication date: 23-Feb-2020
https://dl.acm.org/doi/10.1145/3373087.3375313
Halbiniak KSzustak LOlas TWyrzykowski RGepner P(2020)Exploration of OpenCL Heterogeneous Programming for Porting Solidification Modeling to CPU‐GPU PlatformsConcurrency and Computation: Practice and Experience10.1002/cpe.601133:4Online publication date: 9-Oct-2020
https://doi.org/10.1002/cpe.6011
Davila GOliveira DNavaux PRech P(2019)Identifying the Most Reliable Collaborative Workload Distribution in Heterogeneous Devices2019 Design, Automation & Test in Europe Conference & Exhibition (DATE)10.23919/DATE.2019.8715107(1325-1330)Online publication date: Mar-2019
https://doi.org/10.23919/DATE.2019.8715107
Huang SChang LEl Hajj IGarcia de Gonzalo SGómez-Luna JChalamalasetti SEl-Hadedy MMilojicic DMutlu OChen DHwu WApte VDi Marco ALitoiu MMerseguer J(2019)Analysis and Modeling of Collaborative Execution Strategies for Heterogeneous CPU-FPGA ArchitecturesProceedings of the 2019 ACM/SPEC International Conference on Performance Engineering10.1145/3297663.3310305(79-90)Online publication date: 4-Apr-2019
https://dl.acm.org/doi/10.1145/3297663.3310305
Freytag GSerpa MLima JRech PNavaux P(2019)Non-uniform Partitioning for Collaborative Execution on Heterogeneous Architectures2019 31st International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)10.1109/SBAC-PAD.2019.00031(128-135)Online publication date: Oct-2019
https://doi.org/10.1109/SBAC-PAD.2019.00031
Buyya RSrirama SCasale GCalheiros RSimmhan YVarghese BGelenbe EJavadi BVaquero LNetto MToosi ARodriguez MLlorente IVimercati SSamarati PMilojicic DVarela CBahsoon RAssuncao MRana OZhou WJin HGentzsch WZomaya AShen H(2018)A Manifesto for Future Generation Cloud ComputingACM Computing Surveys10.1145/324173751:5(1-38)Online publication date: 19-Nov-2018
https://dl.acm.org/doi/10.1145/3241737
Sun YMukherjee SBaruah TDong SGutierrez JMohan PKaeli D(2018)Evaluating Performance Tradeoffs on the Radeon Open Compute Platform2018 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)10.1109/ISPASS.2018.00034(209-218)Online publication date: Apr-2018
https://doi.org/10.1109/ISPASS.2018.00034

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Figures

Tables

Media

View Table of Conten