Influences on Throughput and Latency in Stream Programs

Vu Thien Nga Nguyen; Raimund Kirner

Influences on Throughput and Latency in Stream Programs

Vu Thien Nga Nguyen, Raimund Kirner

Centre for Computer Science and Informatics Research

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

28 Downloads (Pure)

Abstract

Stream programming is a promising approach to execute programs on parallel hardware such as multi-core systems. It allows to reuse sequential code at component level and to extend such code with concurrency-handling at the communication level. In this paper we investigate in the performance of stream programs in terms of throughput and latency. We identify factors that affect these performance metrics and propose an efficient scheduling approach to obtain the maximal performance.

Original language	English
Title of host publication	2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures
Number of pages	6
Publication status	Published - 1 Jan 2013
Event	2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures - Berlin, Germany Duration: 22 Jan 2013 → 22 Jan 2016 http://www.project-advance.eu/2012/07/fd-coma-workshop-2013/

Conference

Conference	2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures
Country/Territory	Germany
City	Berlin
Period	22/01/13 → 22/01/16
Internet address	http://www.project-advance.eu/2012/07/fd-coma-workshop-2013/

Keywords

stream processing
performance

Access to Document

rr_2013_FDCOMA13_throughput_latency_streamingLicence: CC BY

Cite this

@inproceedings{46243db9a8fd4d819b719259698ceb16,

title = "Influences on Throughput and Latency in Stream Programs",

abstract = "Stream programming is a promising approach to execute programs on parallel hardware such as multi-core systems. It allows to reuse sequential code at component level and to extend such code with concurrency-handling at the communication level. In this paper we investigate in the performance of stream programs in terms of throughput and latency. We identify factors that affect these performance metrics and propose an efficient scheduling approach to obtain the maximal performance.",

keywords = "stream processing, performance",

author = "Nguyen, {Vu Thien Nga} and Raimund Kirner",

note = "Vu Thien Nga Nguyen and Raimund Kirner, 'Influences on Throughput and Latency in Stream Programs' paper presented at the 2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures. Berlin, Germany. 22 January 2013; 2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures ; Conference date: 22-01-2013 Through 22-01-2016",

year = "2013",

month = jan,

day = "1",

language = "English",

booktitle = "2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures",

url = "http://www.project-advance.eu/2012/07/fd-coma-workshop-2013/",

}

TY - GEN

T1 - Influences on Throughput and Latency in Stream Programs

AU - Nguyen, Vu Thien Nga

AU - Kirner, Raimund

N1 - Vu Thien Nga Nguyen and Raimund Kirner, 'Influences on Throughput and Latency in Stream Programs' paper presented at the 2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures. Berlin, Germany. 22 January 2013

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Stream programming is a promising approach to execute programs on parallel hardware such as multi-core systems. It allows to reuse sequential code at component level and to extend such code with concurrency-handling at the communication level. In this paper we investigate in the performance of stream programs in terms of throughput and latency. We identify factors that affect these performance metrics and propose an efficient scheduling approach to obtain the maximal performance.

AB - Stream programming is a promising approach to execute programs on parallel hardware such as multi-core systems. It allows to reuse sequential code at component level and to extend such code with concurrency-handling at the communication level. In this paper we investigate in the performance of stream programs in terms of throughput and latency. We identify factors that affect these performance metrics and propose an efficient scheduling approach to obtain the maximal performance.

KW - stream processing

KW - performance

M3 - Conference contribution

BT - 2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures

T2 - 2nd Workshop on Feedback-Directed Compiler Optimization for Multi-Core Architectures

Y2 - 22 January 2013 through 22 January 2016

ER -

Influences on Throughput and Latency in Stream Programs

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this