AIWC: OpenCL-Based architecture-independent workload characterization

Beau Johnston; Josh Milthorpe

doi:10.1109/LLVM-HPC.2018.8639381

AIWC: OpenCL-Based architecture-independent workload characterization

Beau Johnston, Josh Milthorpe

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › peer-review

2 Citations (Scopus)

Abstract

Measuring performance-critical characteristics of application workloads is important both for developers, who must understand and optimize the performance of codes, as well as designers and integrators of HPC systems, who must ensure that compute architectures are suitable for the intended workloads. However, if these workload characteristics are tied to architectural features that are specific to a particular system, they may not generalize well to alternative or future systems. An architecture-independent method ensures an accurate characterization of inherent program behaviour, without bias due to architecture-dependent features that vary widely between different types of accelerators. This work presents the first architecture-independent workload characterization framework for heterogeneous compute platforms, proposing a set of metrics determining the suitability and performance of an application on any parallel HPC architecture. The tool, AIWC, is a plugin for the open-source Oclgrind simulator. It supports parallel workloads and is capable of characterizing OpenCL codes currently in use in the supercomputing setting. AIWC simulates an OpenCL device by directly interpreting LLVM instructions, and the resulting metrics may be used for performance prediction and developer feedback to guide device-specific optimizations. An evaluation of the metrics collected over a subset of the Extended OpenDwarfs Benchmark Suite is also presented.

Original language	English
Title of host publication	Proceedings of LLVM-HPC 2018
Subtitle of host publication	5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	81-91
Number of pages	11
ISBN (Electronic)	9781728101880
DOIs	https://doi.org/10.1109/LLVM-HPC.2018.8639381
Publication status	Published - 2 Jul 2018
Event	5th IEEE/ACM Workshop on the LLVM Compiler Infrastructure in HPC, LLVM-HPC 2018 - Dallas, United States Duration: 12 Nov 2018 → …

Publication series

Name	Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	5th IEEE/ACM Workshop on the LLVM Compiler Infrastructure in HPC, LLVM-HPC 2018
Country/Territory	United States
City	Dallas
Period	12/11/18 → …

Access to Document

10.1109/LLVM-HPC.2018.8639381

Cite this

Johnston, B., & Milthorpe, J. (2018). AIWC: OpenCL-Based architecture-independent workload characterization. In Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 81-91). Article 8639381 (Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LLVM-HPC.2018.8639381

Johnston, Beau ; Milthorpe, Josh. / AIWC : OpenCL-Based architecture-independent workload characterization. Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 81-91 (Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{199e4ce2111f4eea83413886b528355e,

title = "AIWC: OpenCL-Based architecture-independent workload characterization",

abstract = "Measuring performance-critical characteristics of application workloads is important both for developers, who must understand and optimize the performance of codes, as well as designers and integrators of HPC systems, who must ensure that compute architectures are suitable for the intended workloads. However, if these workload characteristics are tied to architectural features that are specific to a particular system, they may not generalize well to alternative or future systems. An architecture-independent method ensures an accurate characterization of inherent program behaviour, without bias due to architecture-dependent features that vary widely between different types of accelerators. This work presents the first architecture-independent workload characterization framework for heterogeneous compute platforms, proposing a set of metrics determining the suitability and performance of an application on any parallel HPC architecture. The tool, AIWC, is a plugin for the open-source Oclgrind simulator. It supports parallel workloads and is capable of characterizing OpenCL codes currently in use in the supercomputing setting. AIWC simulates an OpenCL device by directly interpreting LLVM instructions, and the resulting metrics may be used for performance prediction and developer feedback to guide device-specific optimizations. An evaluation of the metrics collected over a subset of the Extended OpenDwarfs Benchmark Suite is also presented.",

keywords = "analysis, workload characterisation",

author = "Beau Johnston and Josh Milthorpe",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 5th IEEE/ACM Workshop on the LLVM Compiler Infrastructure in HPC, LLVM-HPC 2018 ; Conference date: 12-11-2018",

year = "2018",

month = jul,

day = "2",

doi = "10.1109/LLVM-HPC.2018.8639381",

language = "English",

series = "Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "81--91",

booktitle = "Proceedings of LLVM-HPC 2018",

address = "United States",

}

Johnston, B & Milthorpe, J 2018, AIWC: OpenCL-Based architecture-independent workload characterization. in Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis., 8639381, Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 81-91, 5th IEEE/ACM Workshop on the LLVM Compiler Infrastructure in HPC, LLVM-HPC 2018, Dallas, United States, 12/11/18. https://doi.org/10.1109/LLVM-HPC.2018.8639381

AIWC: OpenCL-Based architecture-independent workload characterization. / Johnston, Beau; Milthorpe, Josh.
Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2018. p. 81-91 8639381 (Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › peer-review

TY - GEN

T1 - AIWC

T2 - 5th IEEE/ACM Workshop on the LLVM Compiler Infrastructure in HPC, LLVM-HPC 2018

AU - Johnston, Beau

AU - Milthorpe, Josh

PY - 2018/7/2

Y1 - 2018/7/2

N2 - Measuring performance-critical characteristics of application workloads is important both for developers, who must understand and optimize the performance of codes, as well as designers and integrators of HPC systems, who must ensure that compute architectures are suitable for the intended workloads. However, if these workload characteristics are tied to architectural features that are specific to a particular system, they may not generalize well to alternative or future systems. An architecture-independent method ensures an accurate characterization of inherent program behaviour, without bias due to architecture-dependent features that vary widely between different types of accelerators. This work presents the first architecture-independent workload characterization framework for heterogeneous compute platforms, proposing a set of metrics determining the suitability and performance of an application on any parallel HPC architecture. The tool, AIWC, is a plugin for the open-source Oclgrind simulator. It supports parallel workloads and is capable of characterizing OpenCL codes currently in use in the supercomputing setting. AIWC simulates an OpenCL device by directly interpreting LLVM instructions, and the resulting metrics may be used for performance prediction and developer feedback to guide device-specific optimizations. An evaluation of the metrics collected over a subset of the Extended OpenDwarfs Benchmark Suite is also presented.

AB - Measuring performance-critical characteristics of application workloads is important both for developers, who must understand and optimize the performance of codes, as well as designers and integrators of HPC systems, who must ensure that compute architectures are suitable for the intended workloads. However, if these workload characteristics are tied to architectural features that are specific to a particular system, they may not generalize well to alternative or future systems. An architecture-independent method ensures an accurate characterization of inherent program behaviour, without bias due to architecture-dependent features that vary widely between different types of accelerators. This work presents the first architecture-independent workload characterization framework for heterogeneous compute platforms, proposing a set of metrics determining the suitability and performance of an application on any parallel HPC architecture. The tool, AIWC, is a plugin for the open-source Oclgrind simulator. It supports parallel workloads and is capable of characterizing OpenCL codes currently in use in the supercomputing setting. AIWC simulates an OpenCL device by directly interpreting LLVM instructions, and the resulting metrics may be used for performance prediction and developer feedback to guide device-specific optimizations. An evaluation of the metrics collected over a subset of the Extended OpenDwarfs Benchmark Suite is also presented.

KW - analysis

KW - workload characterisation

UR - http://www.scopus.com/inward/record.url?scp=85063100149&partnerID=8YFLogxK

U2 - 10.1109/LLVM-HPC.2018.8639381

DO - 10.1109/LLVM-HPC.2018.8639381

M3 - Conference Paper

T3 - Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 81

EP - 91

BT - Proceedings of LLVM-HPC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 12 November 2018

ER -

Johnston B, Milthorpe J. AIWC: OpenCL-Based architecture-independent workload characterization. In Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2018. p. 81-91. 8639381. (Proceedings of LLVM-HPC 2018: 5th Workshop on the LLVM Compiler Infrastructure in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/LLVM-HPC.2018.8639381

AIWC: OpenCL-Based architecture-independent workload characterization

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this