Thorsten von Eicken and David E. Culler and Seth Copen Goldstein and Klaus Erik Schauser

EECS Department, University of California, Berkeley

Technical Report No. UCB/CSD-92-675

, 1992

http://www2.eecs.berkeley.edu/Pubs/TechRpts/1992/CSD-92-675.pdf

The design challenge for large-scale multiprocessors is (1) to minimize communication overhead, (2) allow communication to overlap computation, and (3) coordinate the two without sacrificing processor cost/performance. We show that existing message passing multiprocessors have unnecessarily high communication costs. Research prototypes of message driven machines demonstrate low communication overhead, but poor processor cost/performance. We introduce a simple communication mechanism, Active Messages, show that it is intrinsic to both architectures, allows cost effective use of the hardware, and offers tremendous flexibility. Implementations on nCUBE/2 and CM-5 are described and evaluated using a split-phase shared-memory extension to C, Split-C. We further show that active messages are sufficient to implement the dynamically scheduled languages for which message driven machines were designed. With this mechanism, latency tolerance becomes a programming/compiling concern. Hardware support for active messages is desirable and we outline a range of enhancements to mainstream processors.

---

BibTeX citation:

@techreport{von Eicken:CSD-92-675,
    Author= {von Eicken, Thorsten and Culler, David E. and Goldstein, Seth Copen and Schauser, Klaus Erik},
    Title= {Active Messages: a Mechanism for Integrated Communication and Computation},
    Year= {1992},
    Month= {Mar},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/1992/5569.html},
    Number= {UCB/CSD-92-675},
    Abstract= {The design challenge for large-scale multiprocessors is (1) to minimize communication overhead, (2) allow communication to overlap computation, and (3) coordinate the two without sacrificing processor cost/performance. We show that existing message passing multiprocessors have unnecessarily high communication costs. Research prototypes of message driven machines demonstrate low communication overhead, but poor processor cost/performance. We introduce a simple communication mechanism, Active Messages, show that it is intrinsic to both architectures, allows cost effective use of the hardware, and offers tremendous flexibility. Implementations on nCUBE/2 and CM-5 are described and evaluated using a split-phase shared-memory extension to C, Split-C. We further show that active messages are sufficient to implement the dynamically scheduled languages for which message driven machines were designed. With this mechanism, latency tolerance becomes a programming/compiling concern. Hardware support for active messages is desirable and we outline a range of enhancements to mainstream processors.},
}

EndNote citation:

%0 Report
%A von Eicken, Thorsten 
%A Culler, David E. 
%A Goldstein, Seth Copen 
%A Schauser, Klaus Erik 
%T Active Messages: a Mechanism for Integrated Communication and Computation
%I EECS Department, University of California, Berkeley
%D 1992
%@ UCB/CSD-92-675
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/1992/5569.html
%F von Eicken:CSD-92-675