---
_id: '303'
abstract:
- lang: eng
  text: This paper introduces Binary Acceleration At Runtime(BAAR), an easy-to-use
    on-the-fly binary acceleration mechanismwhich aims to tackle the problem of enabling
    existentsoftware to automatically utilize accelerators at runtime. BAARis based
    on the LLVM Compiler Infrastructure and has aclient-server architecture. The client
    runs the program to beaccelerated in an environment which allows program analysisand
    profiling. Program parts which are identified as suitable forthe available accelerator
    are exported and sent to the server.The server optimizes these program parts for
    the acceleratorand provides RPC execution for the client. The client transformsits
    program to utilize accelerated execution on the server foroffloaded program parts.
    We evaluate our work with a proofof-concept implementation of BAAR that uses an
    Intel XeonPhi 5110P as the acceleration target and performs automaticoffloading,
    parallelization and vectorization of suitable programparts. The practicality of
    BAAR for real-world examples is shownbased on a study of stencil codes. Our results
    show a speedup ofup to 4 without any developer-provided hints and 5.77 withhints
    over the same code compiled with the Intel Compiler atoptimization level O2 and
    running on an Intel Xeon E5-2670machine. Based on our insights gained during implementationand
    evaluation we outline future directions of research, e.g.,offloading more fine-granular
    program parts than functions, amore sophisticated communication mechanism or introducing
    onstack-replacement.
author:
- first_name: Marvin
  full_name: Damschen, Marvin
  last_name: Damschen
- first_name: Christian
  full_name: Plessl, Christian
  id: '16153'
  last_name: Plessl
  orcid: 0000-0001-5728-9982
citation:
  ama: 'Damschen M, Plessl C. Easy-to-Use On-The-Fly Binary Program Acceleration on
    Many-Cores. In: <i>Proceedings of the 5th International Workshop on Adaptive Self-Tuning
    Computing Systems (ADAPT)</i>. ; 2015.'
  apa: Damschen, M., &#38; Plessl, C. (2015). Easy-to-Use On-The-Fly Binary Program
    Acceleration on Many-Cores. <i>Proceedings of the 5th International Workshop on
    Adaptive Self-Tuning Computing Systems (ADAPT)</i>.
  bibtex: '@inproceedings{Damschen_Plessl_2015, title={Easy-to-Use On-The-Fly Binary
    Program Acceleration on Many-Cores}, booktitle={Proceedings of the 5th International
    Workshop on Adaptive Self-tuning Computing Systems (ADAPT)}, author={Damschen,
    Marvin and Plessl, Christian}, year={2015} }'
  chicago: Damschen, Marvin, and Christian Plessl. “Easy-to-Use On-The-Fly Binary
    Program Acceleration on Many-Cores.” In <i>Proceedings of the 5th International
    Workshop on Adaptive Self-Tuning Computing Systems (ADAPT)</i>, 2015.
  ieee: M. Damschen and C. Plessl, “Easy-to-Use On-The-Fly Binary Program Acceleration
    on Many-Cores,” 2015.
  mla: Damschen, Marvin, and Christian Plessl. “Easy-to-Use On-The-Fly Binary Program
    Acceleration on Many-Cores.” <i>Proceedings of the 5th International Workshop
    on Adaptive Self-Tuning Computing Systems (ADAPT)</i>, 2015.
  short: 'M. Damschen, C. Plessl, in: Proceedings of the 5th International Workshop
    on Adaptive Self-Tuning Computing Systems (ADAPT), 2015.'
date_created: 2017-10-17T12:41:51Z
date_updated: 2023-09-26T13:29:59Z
ddc:
- '040'
department:
- _id: '27'
- _id: '518'
- _id: '78'
external_id:
  arxiv:
  - '1412.3906'
file:
- access_level: open_access
  content_type: application/pdf
  creator: florida
  date_created: 2018-03-20T07:46:46Z
  date_updated: 2019-08-01T09:10:44Z
  file_id: '1442'
  file_name: 303-plessl15_adapt.pdf
  file_size: 1176620
  relation: main_file
file_date_updated: 2019-08-01T09:10:44Z
has_accepted_license: '1'
language:
- iso: eng
oa: '1'
project:
- _id: '1'
  grant_number: '160364472'
  name: SFB 901
- _id: '14'
  grant_number: '160364472'
  name: SFB 901 - Subprojekt C2
- _id: '4'
  name: SFB 901 - Project Area C
- _id: '34'
  grant_number: '610996'
  name: Self-Adaptive Virtualisation-Aware High-Performance/Low-Energy Heterogeneous
    System Architectures
publication: Proceedings of the 5th International Workshop on Adaptive Self-tuning
  Computing Systems (ADAPT)
quality_controlled: '1'
status: public
title: Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores
type: conference
user_id: '15278'
year: '2015'
...