ACM Home Page
Please provide us with feedback. Feedback
Performance benefits of large execution atomic units in dynamically scheduled machines
Full text PdfPdf (672 KB)
Source International Conference on Supercomputing archive
Proceedings of the 3rd international conference on Supercomputing table of contents
Crete, Greece
Pages: 427 - 432  
Year of Publication: 1989
ISBN:0-89791-309-4
Authors
Stephen W. Melvin  Computer Science Division, University of California, Berkeley, CA
Yale N. Patt  Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI
Sponsors
Computer Tech Inst. : Computer Technology Institute
SIGARCH: ACM Special Interest Group on Computer Architecture
SIAM : Society for Industrial and Applied Mathematics
AICA : Assoc Italianai de Calcolo Automatico
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 9,   Downloads (12 Months): 22,   Citation Count: 8
Additional Information:

abstract   references   cited by   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/318789.318890
What is a DOI?

ABSTRACT

In this paper we identify three types of atomic units, or indivisible units of work: architectural atomic units (defined by architecture level interrupts and exceptions), compiler atomic units (defined by compiler code generation) and execution atomic units (defined by run-time interruptibility). We discuss trade-offs for these units and show that size has different performance implications depending on the atomic unit. We simulate a number of different implementations of the VAX architecture, focusing on different execution atomic unit sizes. We show that significant performance benefits can be achieved by having large execution atomic units in dynamically scheduled machines.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
J. A. Fisher, "Trace Scheduling' ATe, l,,t,que for Global Microcode Compaction," {EEl"_.' Transactions on Computers, vo{, 0-30, rio. 7, July 1981.
 
2
3
 
4
5
 
6
Yale N. P~tt, Michael C. Shebanow. Wen-reel Hwu and Stephen W. Melvin, "A C Conlpiler for HPS 1, A Highly ParMle{ Execution Engine," Proceedings, 19tb Hawaii fnternational Conferemce on System Sciences, Honolulu, HI..january 1986.
 
7
George R~ai-, "The 801 Minicomputer," IBM JournM of Research and Development. Vol. 27, No. 3, May 1983, pp. 2aZ-'246.
 
8
R. M. Tomas.ulo, "An Efficient Algorithm fo~ Exploiting Multiple Arithmetic Units," fBM Journal of Research and Development, Vol. 11, 1967, pp. 25-33.

CITED BY  8

Collaborative Colleagues:
Stephen W. Melvin: colleagues
Yale N. Patt: colleagues