moskowit@paul.rutgers.edu (Len Moskowitz) (08/16/90)
Sun/Transputer Benchmarks for CLIPS Production Systems
DISCUSSION
Back in June and July I posted a few queries about small scale parallel
architectures (less than 64 computing nodes) for hosting production system
languages (e.g., CMU's OPS5 and NASA's CLIPS). Since that time we've narrowed
our choice of microprocessor to the Inmos Transputers (the T800 and T801) and
the Motorola 88000 RISC. One of our constraints is that the final system must
occupy no more that three 6U VME slots. Both Inmos and Transtech have 6U
boards (the B014 and the MCP500 respectively) that accommodate eight size-1
transputer modules ("TRAMS"). Tadpole Technologies has a three board set base
on Motorola's Hypermodule that hosts eight 88100 CPUs and sixteen 88200
cache/MMUs.
We decided to initially benchmark single processor configurations of the
Transputer and 88000 versus our Sun development system and embedded target
hardware. At the moment we have an Inmos Sun 3-based transputer development
system for evaluation. It has a single 25 megaHertz T800 size-2 B404-5A
Transputer Module (TRAM) on a B014-1B and was installed in a Sun 3/80 with a
9U-to-6U adapter. Installation, including rebuilding the kernel, took less
than two hours. Using Inmos's C compiler it took less than 6 hours to port
CLIPS 4.3 from the Suns to the Transputer and to develop the benchmark
applications. This is a tribute to both CLIPS' portability and the efficiency
of the Inmos tools.
Our application programs are typically developed on Suns and ported to a
Motorola MVME-147 single board computer that is embedded in our products. The
-147 uses a 25 megaHertz 68030 and 68882. The Sun 3/80 uses a 20 megaHertz
68030 and 68882 floating point coprocessor. The B404-5A uses a 25 megaHertz
T800-25 Transputer with an integral floating point unit. The -147 runs VxWorks
4.0, a real time UNIX kernel. The Suns run SunOS 4.1. The -147 uses 70
nanosecond, 1 wait state memory. We don't know the memory parameters for the
Sun. The Transputer used 32 kilobytes of 0 wait state SRAM and 2 megabytes of
1 (sometimes 2) wait state DRAM. Both the Sun and the -147 run code generated
by the same Sun C compiler, albeit linked with different operating system
service libraries. The Transputer used code generated by the Inmos C compiler.
All processes on the -147 other than the benchmark were suspended (these
suspended processes occupied no more than one percent of the processing
capacity of the -147). The Suns usual processes and their priorities weren't
interfered with in any way except that remote logins were disabled. The
Transputer ran only the benchmark program.
The benchmark, an expert system for waveform analysis consisting of 71 rules
averaging 4 condition elements and 4 actions, performed no output to the
display and required no disk access. All data required for execution (the
working memory elements) were preloaded into working memory. The rules were
pre-compiled. Timing results shown below are the average of execution for 20
different data sets. Timing ratios (Transputer versus Sun, and Transputer
versus -147) were amazingly constant for the 20 data sets, varying less than
two percent. Timings were measured from the invocation of the production
system until rule firing ceased.
RESULTS
The Transputer was 2.24 and 1.48 times faster than the Sun 3/80 and the -147
board respectively. The -147 board was uniformly about fifty percent faster
than the Sun 3/80.
CREDITS
The benchmarks were designed and run by Dr. Elliott Greene at Bendix Test
Systems Division, Teterboro, New Jersey, 07608, USA.
Bob Kirk from Inmos's Columbia, Maryland office assisted us when we hit snags.
We thank him for his informed and enthusiastic support.
NEXT:
We will have a Tadpole 88000 board in for evaluation the last week in August.
I hope to report on its relative performance shortly thereafter.
Len Moskowitz
Allied-Signal Aerospace Co.
moskowitz@bendix.com (CSnet) Bendix Test Systems Division
moskowitz%bendix.com@relay.cs.net (ARPAnet) Mail code 4/8
...rutgers!bendix.com!moskowitz Teterboro, NJ 07608
moskowit@paul.rutgers.edu