Compilation techniques for out-of-core parallel computations

M. Kandemir; A. Choudhary; J. Ramanujam; R. Bordawekar

doi:10.1016/s0167-8191(98)00027-1

Compilation techniques for out-of-core parallel computations

M. Kandemir^*, A. Choudhary, J. Ramanujam, R. Bordawekar

^*Corresponding author for this work

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

19 Scopus citations

Abstract

The difficulty of handling out-of-core data limits the performance of supercomputers as well as the potential of the parallel machines. Since writing an efficient out-of-core version of a program is a difficult task and virtual memory systems do not perform well on scientific computations, we believe that there is a clear need for compiler directed explicit I/O approach for out-of-core computations. In this paper, we first present an out-of-core compilation strategy based on a disk storage abstraction. Then, we offer a compiler algorithm to optimize locality of disk accesses in out-of-core codes by choosing a good combination of file layouts on disks and loop transformations. We introduce memory coefficient and processor coefficient concepts to characterize the behavior of out-of-core programs under different memory constraints. We also enhance our algorithm to handle data-parallel programs which contain multiple loop nest. Our initial experimental results obtained on IBM SP-2 and Intel Paragon provide encouraging evidence that our approach is successful at optimizing programs which depend on disk-resident data in distributed-memory machines.

Original language	English (US)
Pages (from-to)	597-628
Number of pages	32
Journal	Parallel Computing
Volume	24
Issue number	3-4
DOIs	https://doi.org/10.1016/s0167-8191(98)00027-1
State	Published - May 1998

Funding

This work was supported in part by NSF Young Investigator Award CCR-9357840, NSF CCR-9509143 and in part by the Scalable I/O Initiative, contract number DABT63-94-C-0049 from Defense Advanced Research Projects Agency (DARPA) administered by US Army at Fort Huachuca. The work of J. Ramanujam was supported in part by an NSF Young Investigator Award CCR-9457768, an NSF grant CCR-9210422 and by the Louisiana Board of Regents through contract LEQSF (1991–1994)-RD-A-09.

Keywords

Data-parallelism
Distributed-memory machines
I/O
Optimizing compilers
Out-of-core

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computer Networks and Communications
Computer Graphics and Computer-Aided Design
Artificial Intelligence

Access to Document

10.1016/s0167-8191(98)00027-1

Cite this

@article{a4bf7583aefc4956ab12a8b57c703511,

title = "Compilation techniques for out-of-core parallel computations",

abstract = "The difficulty of handling out-of-core data limits the performance of supercomputers as well as the potential of the parallel machines. Since writing an efficient out-of-core version of a program is a difficult task and virtual memory systems do not perform well on scientific computations, we believe that there is a clear need for compiler directed explicit I/O approach for out-of-core computations. In this paper, we first present an out-of-core compilation strategy based on a disk storage abstraction. Then, we offer a compiler algorithm to optimize locality of disk accesses in out-of-core codes by choosing a good combination of file layouts on disks and loop transformations. We introduce memory coefficient and processor coefficient concepts to characterize the behavior of out-of-core programs under different memory constraints. We also enhance our algorithm to handle data-parallel programs which contain multiple loop nest. Our initial experimental results obtained on IBM SP-2 and Intel Paragon provide encouraging evidence that our approach is successful at optimizing programs which depend on disk-resident data in distributed-memory machines.",

keywords = "Data-parallelism, Distributed-memory machines, I/O, Optimizing compilers, Out-of-core",

author = "M. Kandemir and A. Choudhary and J. Ramanujam and R. Bordawekar",

note = "Funding Information: This work was supported in part by NSF Young Investigator Award CCR-9357840, NSF CCR-9509143 and in part by the Scalable I/O Initiative, contract number DABT63-94-C-0049 from Defense Advanced Research Projects Agency (DARPA) administered by US Army at Fort Huachuca. The work of J. Ramanujam was supported in part by an NSF Young Investigator Award CCR-9457768, an NSF grant CCR-9210422 and by the Louisiana Board of Regents through contract LEQSF (1991–1994)-RD-A-09. ",

year = "1998",

month = may,

doi = "10.1016/s0167-8191(98)00027-1",

language = "English (US)",

volume = "24",

pages = "597--628",

journal = "Parallel Computing",

issn = "0167-8191",

publisher = "Elsevier B.V.",

number = "3-4",

}

TY - JOUR

T1 - Compilation techniques for out-of-core parallel computations

AU - Kandemir, M.

AU - Choudhary, A.

AU - Ramanujam, J.

AU - Bordawekar, R.

N1 - Funding Information: This work was supported in part by NSF Young Investigator Award CCR-9357840, NSF CCR-9509143 and in part by the Scalable I/O Initiative, contract number DABT63-94-C-0049 from Defense Advanced Research Projects Agency (DARPA) administered by US Army at Fort Huachuca. The work of J. Ramanujam was supported in part by an NSF Young Investigator Award CCR-9457768, an NSF grant CCR-9210422 and by the Louisiana Board of Regents through contract LEQSF (1991–1994)-RD-A-09.

PY - 1998/5

Y1 - 1998/5

N2 - The difficulty of handling out-of-core data limits the performance of supercomputers as well as the potential of the parallel machines. Since writing an efficient out-of-core version of a program is a difficult task and virtual memory systems do not perform well on scientific computations, we believe that there is a clear need for compiler directed explicit I/O approach for out-of-core computations. In this paper, we first present an out-of-core compilation strategy based on a disk storage abstraction. Then, we offer a compiler algorithm to optimize locality of disk accesses in out-of-core codes by choosing a good combination of file layouts on disks and loop transformations. We introduce memory coefficient and processor coefficient concepts to characterize the behavior of out-of-core programs under different memory constraints. We also enhance our algorithm to handle data-parallel programs which contain multiple loop nest. Our initial experimental results obtained on IBM SP-2 and Intel Paragon provide encouraging evidence that our approach is successful at optimizing programs which depend on disk-resident data in distributed-memory machines.

AB - The difficulty of handling out-of-core data limits the performance of supercomputers as well as the potential of the parallel machines. Since writing an efficient out-of-core version of a program is a difficult task and virtual memory systems do not perform well on scientific computations, we believe that there is a clear need for compiler directed explicit I/O approach for out-of-core computations. In this paper, we first present an out-of-core compilation strategy based on a disk storage abstraction. Then, we offer a compiler algorithm to optimize locality of disk accesses in out-of-core codes by choosing a good combination of file layouts on disks and loop transformations. We introduce memory coefficient and processor coefficient concepts to characterize the behavior of out-of-core programs under different memory constraints. We also enhance our algorithm to handle data-parallel programs which contain multiple loop nest. Our initial experimental results obtained on IBM SP-2 and Intel Paragon provide encouraging evidence that our approach is successful at optimizing programs which depend on disk-resident data in distributed-memory machines.

KW - Data-parallelism

KW - Distributed-memory machines

KW - I/O

KW - Optimizing compilers

KW - Out-of-core

UR - http://www.scopus.com/inward/record.url?scp=0032066688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032066688&partnerID=8YFLogxK

U2 - 10.1016/s0167-8191(98)00027-1

DO - 10.1016/s0167-8191(98)00027-1

M3 - Article

AN - SCOPUS:0032066688

SN - 0167-8191

VL - 24

SP - 597

EP - 628

JO - Parallel Computing

JF - Parallel Computing

IS - 3-4

ER -

Compilation techniques for out-of-core parallel computations

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this