Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster

Khadeer Ahmed, Qinru Qiu, Mangesh Tamhankar

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

Abstract

With the proliferation of application specific accelerators, the use of heterogeneous clusters is rapidly increasing. Consisting of processors with different architectures, a heterogeneous cluster aims at providing different performance and cost tradeoffs for different types of workloads. In order to achieve peak performance, software running on heterogeneous cluster needs to be designed carefully to provide enough flexibility to explore its variety. We propose a design methodology to modularize complex software applications with data dependencies. The software application designed in this way have the flexibility to be reconfigured for different hardware platforms to facilitate resource management, and features high scalability and parallelism. Using a neuromorphic application as a case study, we present the concept of modularization and discuss the management, scheduling and communication of the modules. We also present experimental results demonstrating the improvements and effects of system scaling on throughput.

Original languageEnglish (US)
Title of host publication2016 IEEE High Performance Extreme Computing Conference, HPEC 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509035250
DOIs
StatePublished - Nov 28 2016
Event2016 IEEE High Performance Extreme Computing Conference, HPEC 2016 - Waltham, United States
Duration: Sep 13 2016Sep 15 2016

Other

Other2016 IEEE High Performance Extreme Computing Conference, HPEC 2016
CountryUnited States
CityWaltham
Period9/13/169/15/16

Fingerprint

Application programs
Particle accelerators
Scalability
Flexibility
Scheduling
Throughput
Software Performance
Hardware
Data Dependency
Modularization
Software
Communication
Resource Management
Accelerator
Proliferation
Design Methodology
Parallelism
Workload
Costs
Trade-offs

Keywords

  • Distributed computing
  • heterogeneous computing
  • latency hiding
  • modularization
  • pipelining
  • structure based scheduling

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Hardware and Architecture
  • Computational Mathematics

Cite this

Ahmed, K., Qiu, Q., & Tamhankar, M. (2016). Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster. In 2016 IEEE High Performance Extreme Computing Conference, HPEC 2016 [7761598] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/HPEC.2016.7761598

Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster. / Ahmed, Khadeer; Qiu, Qinru; Tamhankar, Mangesh.

2016 IEEE High Performance Extreme Computing Conference, HPEC 2016. Institute of Electrical and Electronics Engineers Inc., 2016. 7761598.

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

Ahmed, K, Qiu, Q & Tamhankar, M 2016, Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster. in 2016 IEEE High Performance Extreme Computing Conference, HPEC 2016., 7761598, Institute of Electrical and Electronics Engineers Inc., 2016 IEEE High Performance Extreme Computing Conference, HPEC 2016, Waltham, United States, 9/13/16. https://doi.org/10.1109/HPEC.2016.7761598
Ahmed K, Qiu Q, Tamhankar M. Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster. In 2016 IEEE High Performance Extreme Computing Conference, HPEC 2016. Institute of Electrical and Electronics Engineers Inc. 2016. 7761598 https://doi.org/10.1109/HPEC.2016.7761598
Ahmed, Khadeer ; Qiu, Qinru ; Tamhankar, Mangesh. / Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster. 2016 IEEE High Performance Extreme Computing Conference, HPEC 2016. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{303d4efeac9c4b109335a924595d6a56,
title = "Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster",
abstract = "With the proliferation of application specific accelerators, the use of heterogeneous clusters is rapidly increasing. Consisting of processors with different architectures, a heterogeneous cluster aims at providing different performance and cost tradeoffs for different types of workloads. In order to achieve peak performance, software running on heterogeneous cluster needs to be designed carefully to provide enough flexibility to explore its variety. We propose a design methodology to modularize complex software applications with data dependencies. The software application designed in this way have the flexibility to be reconfigured for different hardware platforms to facilitate resource management, and features high scalability and parallelism. Using a neuromorphic application as a case study, we present the concept of modularization and discuss the management, scheduling and communication of the modules. We also present experimental results demonstrating the improvements and effects of system scaling on throughput.",
keywords = "Distributed computing, heterogeneous computing, latency hiding, modularization, pipelining, structure based scheduling",
author = "Khadeer Ahmed and Qinru Qiu and Mangesh Tamhankar",
year = "2016",
month = "11",
day = "28",
doi = "10.1109/HPEC.2016.7761598",
language = "English (US)",
booktitle = "2016 IEEE High Performance Extreme Computing Conference, HPEC 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Distributed and configurable architecture for neuromorphic applications on heterogeneous cluster

AU - Ahmed, Khadeer

AU - Qiu, Qinru

AU - Tamhankar, Mangesh

PY - 2016/11/28

Y1 - 2016/11/28

N2 - With the proliferation of application specific accelerators, the use of heterogeneous clusters is rapidly increasing. Consisting of processors with different architectures, a heterogeneous cluster aims at providing different performance and cost tradeoffs for different types of workloads. In order to achieve peak performance, software running on heterogeneous cluster needs to be designed carefully to provide enough flexibility to explore its variety. We propose a design methodology to modularize complex software applications with data dependencies. The software application designed in this way have the flexibility to be reconfigured for different hardware platforms to facilitate resource management, and features high scalability and parallelism. Using a neuromorphic application as a case study, we present the concept of modularization and discuss the management, scheduling and communication of the modules. We also present experimental results demonstrating the improvements and effects of system scaling on throughput.

AB - With the proliferation of application specific accelerators, the use of heterogeneous clusters is rapidly increasing. Consisting of processors with different architectures, a heterogeneous cluster aims at providing different performance and cost tradeoffs for different types of workloads. In order to achieve peak performance, software running on heterogeneous cluster needs to be designed carefully to provide enough flexibility to explore its variety. We propose a design methodology to modularize complex software applications with data dependencies. The software application designed in this way have the flexibility to be reconfigured for different hardware platforms to facilitate resource management, and features high scalability and parallelism. Using a neuromorphic application as a case study, we present the concept of modularization and discuss the management, scheduling and communication of the modules. We also present experimental results demonstrating the improvements and effects of system scaling on throughput.

KW - Distributed computing

KW - heterogeneous computing

KW - latency hiding

KW - modularization

KW - pipelining

KW - structure based scheduling

UR - http://www.scopus.com/inward/record.url?scp=85007107650&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85007107650&partnerID=8YFLogxK

U2 - 10.1109/HPEC.2016.7761598

DO - 10.1109/HPEC.2016.7761598

M3 - Conference contribution

AN - SCOPUS:85007107650

BT - 2016 IEEE High Performance Extreme Computing Conference, HPEC 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -