Orchestration by approximation

S. M. Farhad, Yousun Ko, bernd Burgstaller, Bernhard Scholz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

17 Citations (Scopus)

Abstract

We present a novel 2-approximation algorithm for deploying stream graphs on multicore computers and a stream graph transformation that eliminates bottlenecks. The key technical insight is a data rate transfer model that enables the computation of a "closed form", i.e., the data rate transfer function of an actor depending on the arrival rate of the stream program. A combinatorial optimization problem uses the closed form to maximize the throughput of the stream program. Although the problem is inherently NP-hard, we present an efficient and effective 2-approximation algorithm that provides a lower bound on the quality of the solution. We introduce a transformation that uses the closed form to identify and eliminate bottlenecks. We show experimentally that state-of-the art integer linear programming approaches for orchestrating stream graphs are (1) intractable or at least impractical for larger stream graphs and larger number of processors and (2) our 2-approximation algorithm is highly efficient and its results are close to the optimal solution for a standard set of StreamIt benchmark programs.

Original languageEnglish
Title of host publicationASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems
Pages357-367
Number of pages11
DOIs
Publication statusPublished - 2011 Mar 31
Event16th International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2011 - Newport Beach, CA, United States
Duration: 2011 Mar 52011 Mar 11

Publication series

NameInternational Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS

Other

Other16th International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2011
CountryUnited States
CityNewport Beach, CA
Period11/3/511/3/11

Fingerprint

Approximation algorithms
Data transfer rates
Combinatorial optimization
Linear programming
Transfer functions
Throughput

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems
  • Hardware and Architecture

Cite this

Farhad, S. M., Ko, Y., Burgstaller, B., & Scholz, B. (2011). Orchestration by approximation. In ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems (pp. 357-367). (International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS). https://doi.org/10.1145/1950365.1950406
Farhad, S. M. ; Ko, Yousun ; Burgstaller, bernd ; Scholz, Bernhard. / Orchestration by approximation. ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems. 2011. pp. 357-367 (International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS).
@inproceedings{39b4909e7cb84db8ba868bde3ec655e3,
title = "Orchestration by approximation",
abstract = "We present a novel 2-approximation algorithm for deploying stream graphs on multicore computers and a stream graph transformation that eliminates bottlenecks. The key technical insight is a data rate transfer model that enables the computation of a {"}closed form{"}, i.e., the data rate transfer function of an actor depending on the arrival rate of the stream program. A combinatorial optimization problem uses the closed form to maximize the throughput of the stream program. Although the problem is inherently NP-hard, we present an efficient and effective 2-approximation algorithm that provides a lower bound on the quality of the solution. We introduce a transformation that uses the closed form to identify and eliminate bottlenecks. We show experimentally that state-of-the art integer linear programming approaches for orchestrating stream graphs are (1) intractable or at least impractical for larger stream graphs and larger number of processors and (2) our 2-approximation algorithm is highly efficient and its results are close to the optimal solution for a standard set of StreamIt benchmark programs.",
author = "Farhad, {S. M.} and Yousun Ko and bernd Burgstaller and Bernhard Scholz",
year = "2011",
month = "3",
day = "31",
doi = "10.1145/1950365.1950406",
language = "English",
isbn = "9781450302661",
series = "International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS",
pages = "357--367",
booktitle = "ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems",

}

Farhad, SM, Ko, Y, Burgstaller, B & Scholz, B 2011, Orchestration by approximation. in ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems. International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS, pp. 357-367, 16th International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2011, Newport Beach, CA, United States, 11/3/5. https://doi.org/10.1145/1950365.1950406

Orchestration by approximation. / Farhad, S. M.; Ko, Yousun; Burgstaller, bernd; Scholz, Bernhard.

ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems. 2011. p. 357-367 (International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Orchestration by approximation

AU - Farhad, S. M.

AU - Ko, Yousun

AU - Burgstaller, bernd

AU - Scholz, Bernhard

PY - 2011/3/31

Y1 - 2011/3/31

N2 - We present a novel 2-approximation algorithm for deploying stream graphs on multicore computers and a stream graph transformation that eliminates bottlenecks. The key technical insight is a data rate transfer model that enables the computation of a "closed form", i.e., the data rate transfer function of an actor depending on the arrival rate of the stream program. A combinatorial optimization problem uses the closed form to maximize the throughput of the stream program. Although the problem is inherently NP-hard, we present an efficient and effective 2-approximation algorithm that provides a lower bound on the quality of the solution. We introduce a transformation that uses the closed form to identify and eliminate bottlenecks. We show experimentally that state-of-the art integer linear programming approaches for orchestrating stream graphs are (1) intractable or at least impractical for larger stream graphs and larger number of processors and (2) our 2-approximation algorithm is highly efficient and its results are close to the optimal solution for a standard set of StreamIt benchmark programs.

AB - We present a novel 2-approximation algorithm for deploying stream graphs on multicore computers and a stream graph transformation that eliminates bottlenecks. The key technical insight is a data rate transfer model that enables the computation of a "closed form", i.e., the data rate transfer function of an actor depending on the arrival rate of the stream program. A combinatorial optimization problem uses the closed form to maximize the throughput of the stream program. Although the problem is inherently NP-hard, we present an efficient and effective 2-approximation algorithm that provides a lower bound on the quality of the solution. We introduce a transformation that uses the closed form to identify and eliminate bottlenecks. We show experimentally that state-of-the art integer linear programming approaches for orchestrating stream graphs are (1) intractable or at least impractical for larger stream graphs and larger number of processors and (2) our 2-approximation algorithm is highly efficient and its results are close to the optimal solution for a standard set of StreamIt benchmark programs.

UR - http://www.scopus.com/inward/record.url?scp=79953120058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79953120058&partnerID=8YFLogxK

U2 - 10.1145/1950365.1950406

DO - 10.1145/1950365.1950406

M3 - Conference contribution

SN - 9781450302661

T3 - International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS

SP - 357

EP - 367

BT - ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems

ER -

Farhad SM, Ko Y, Burgstaller B, Scholz B. Orchestration by approximation. In ASPLOS XVI - 16th International Conference on Architectural Support for Programming Languages and Operating Systems. 2011. p. 357-367. (International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS). https://doi.org/10.1145/1950365.1950406