Shared scratch pad memory space management across applications

Ozcan Ozturk, Mahmut Kandemir, Seung Woo Son, Ibrahim Kolcu

Research output: Contribution to journalArticle

Abstract

Scratch Pad Memories (SPMs) have received considerable attention lately as on-chip memory building blocks. The main characteristic that distinguishes an SPM from a conventional cache memory is that the data flow is controlled by software. The main focus of this paper is the management of an SPM space shared by multiple applications that can potentially share data. The proposed approach has three major components; a compiler analysis phase, a runtime space partitioner, and a local partitioning phase. Our experimental results show that the proposed approach leads to minimum completion time among all alternate memory partitioning schemes tested.

Original languageEnglish (US)
Pages (from-to)54-65
Number of pages12
JournalInternational Journal of Embedded Systems
Volume4
Issue number1
DOIs
StatePublished - Aug 3 2009

Fingerprint

Data storage equipment
Cache memory

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture

Cite this

Ozturk, Ozcan ; Kandemir, Mahmut ; Son, Seung Woo ; Kolcu, Ibrahim. / Shared scratch pad memory space management across applications. In: International Journal of Embedded Systems. 2009 ; Vol. 4, No. 1. pp. 54-65.
@article{34d9b8c434844c1a958f89ce4091d839,
title = "Shared scratch pad memory space management across applications",
abstract = "Scratch Pad Memories (SPMs) have received considerable attention lately as on-chip memory building blocks. The main characteristic that distinguishes an SPM from a conventional cache memory is that the data flow is controlled by software. The main focus of this paper is the management of an SPM space shared by multiple applications that can potentially share data. The proposed approach has three major components; a compiler analysis phase, a runtime space partitioner, and a local partitioning phase. Our experimental results show that the proposed approach leads to minimum completion time among all alternate memory partitioning schemes tested.",
author = "Ozcan Ozturk and Mahmut Kandemir and Son, {Seung Woo} and Ibrahim Kolcu",
year = "2009",
month = "8",
day = "3",
doi = "10.1504/IJES.2009.027240",
language = "English (US)",
volume = "4",
pages = "54--65",
journal = "International Journal of Embedded Systems",
issn = "1741-1068",
publisher = "Inderscience Enterprises Ltd",
number = "1",

}

Shared scratch pad memory space management across applications. / Ozturk, Ozcan; Kandemir, Mahmut; Son, Seung Woo; Kolcu, Ibrahim.

In: International Journal of Embedded Systems, Vol. 4, No. 1, 03.08.2009, p. 54-65.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Shared scratch pad memory space management across applications

AU - Ozturk, Ozcan

AU - Kandemir, Mahmut

AU - Son, Seung Woo

AU - Kolcu, Ibrahim

PY - 2009/8/3

Y1 - 2009/8/3

N2 - Scratch Pad Memories (SPMs) have received considerable attention lately as on-chip memory building blocks. The main characteristic that distinguishes an SPM from a conventional cache memory is that the data flow is controlled by software. The main focus of this paper is the management of an SPM space shared by multiple applications that can potentially share data. The proposed approach has three major components; a compiler analysis phase, a runtime space partitioner, and a local partitioning phase. Our experimental results show that the proposed approach leads to minimum completion time among all alternate memory partitioning schemes tested.

AB - Scratch Pad Memories (SPMs) have received considerable attention lately as on-chip memory building blocks. The main characteristic that distinguishes an SPM from a conventional cache memory is that the data flow is controlled by software. The main focus of this paper is the management of an SPM space shared by multiple applications that can potentially share data. The proposed approach has three major components; a compiler analysis phase, a runtime space partitioner, and a local partitioning phase. Our experimental results show that the proposed approach leads to minimum completion time among all alternate memory partitioning schemes tested.

UR - http://www.scopus.com/inward/record.url?scp=67749111580&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67749111580&partnerID=8YFLogxK

U2 - 10.1504/IJES.2009.027240

DO - 10.1504/IJES.2009.027240

M3 - Article

AN - SCOPUS:67749111580

VL - 4

SP - 54

EP - 65

JO - International Journal of Embedded Systems

JF - International Journal of Embedded Systems

SN - 1741-1068

IS - 1

ER -