From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-0.8 required=5.0 tests=BAYES_00,INVALID_DATE
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,ac5ea38cace9d807,start
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 1993-03-31 17:00:16 PST
Newsgroups: comp.lang.ada
Path: sparky!uunet!gatech!usenet.ins.cwru.edu!cert.org!news.sei.cmu.edu!na
From: na@sei.cmu.edu (Neal Altman)
Subject: Ada Candidates for the SPEC Benchmark Suite
Message-ID: <1993Mar31.192627.20600@sei.cmu.edu>
Keywords: SPEC, Ada, benchmarks
Sender: netnews@sei.cmu.edu (Netnews)
Organization: The Software Engineering Institute
Date: Wed, 31 Mar 1993 19:26:27 EST
Date: 1993-03-31T19:26:27-05:00
List-Id: <comp.lang.ada>


Ada Benchmarks for SPEC
-----------------------

Under the sponsorship of the Ada Joint Program office, the Real-Time
Embedded Systems Testbed (REST) Project at the Software Engineering
Institute is soliciting candidate Ada programs for inclusion as part of
the SPEC benchmark suites.

The Standard Performance Evaluation Corporation (SPEC) is a non-profit
corporation formed to establish, maintain and endorse a standardized set
of relevant benchmarks for high-performance computers. Currently SPEC
benchmark suites include:

      o	CINT92 -- 6 integer intensive execution benchmarks
      o	CFP92 -- 14 floating point intensive execution benchmarks
      o	SDM -- 2 benchmarks of system support for software development

SPEC benchmarks are currently written in FORTRAN and C. The benchmark
suites also include a harness and utilities for running the benchmarks,
measuring times, and computing results.

Ideal Ada candidates for the CINT/CFP benchmark suites should:

      o	Be real applications of substantial size
      o	Be portable across many systems and Ada compilers
      o	Put stress on high-performance systems
      o	Execute for at least 3000 seconds on a VAX11/780
      o	Allow memory usage and instruction mix to be measured
      o	Be capable of performing self checking of results

Ideal Ada candidates for the SDM series of benchmarks should:

      o	Represent Ada software development scenarios
      o	Be portable across many systems and Ada compilers
      o	Stress system software, particularly the compiler and library
      o	Allow multiple copies to execute at the same time 
      o	Not require human input

After testing, selected Ada benchmarks will be submitted to SPEC for
review and possible inclusion in the next release of the SPEC benchmark
suites.

To submit benchmarks or for more information, contact:

	Neal Altman
	Software Engineering Institute
	Carnegie Mellon University
	Pittsburgh, PA 15213
	USA

	Telephone: (412) 268-7613
	FAX: (412) 268-5758
	Internet: na@sei.cmu.edu