Albert D. Cahalan writes:
: The lack of choice you advocate doesn't allow very much thought.
Albert, suppose there was an area where you were well versed. Suppose
that you knew that it was a fruitless area of research. Wouldn't you
feel some obligation to steer new people away from that area? Or would
you just laugh up your sleeve?
: The cheapest 20-way SMP system I could find was a Sun box with
: 400-MHz SPARC processors and 20 GB RAM. It was $564000 w/o drives.
:
: Penguin Computing: Two 550 MHz Pentium III, 256 MB, 13.6 GB for $2115
: Penguin Computing: AMD Athlon 650 MHz, 256 MB, 13.6 GB for $1815
: No-name: AMD Athlon 600, 128 MB, 8.4 GB, "Cool Linux Keyboard" for $939
: Microway: 533 MHz Alpha 21164, 64 MB, 6.5 GB for $1995
:
: Hmmm, 564 k$ for SMP. Roughly 200 to 400 k$ for the cluster.
: Not that this isn't an insane comparison... hope you like Solaris.
How about a little truth in advertising, eh? I'm not a huge fan of Suns
these days but you are painting an extremely misleading picture.
SMP Cluster
bandwidth ~1GB/sec ~.01GB/sec
latency ~200ns ~150,000ns if using messages,
~2,000,000ns if using DSM
It's very common in clusters to spend as much on the interconnect as all
the hardware and software put together. So double your costs and you
still have lower bandwidth and dramatically higher latency.
: >> It all depends on how you define "shared" and "network" I suppose,
: >> but I'm certainly not using SMP hardware or TCP/IP. I can define
: >> physical address spaces that, when accessed, cause automatic data
: >> movement with hardware routing at hundreds of megabytes/second.
: >> (note BYTES not BITS) I think DSM would be perfect here.
: >
: > Very nice. What's the latency, though?
:
: This all depends on the exact hardware version, and I doubt I'm
: allowed to tell you the latest numbers. Generally though, latency
: is lower for shared memory. It depends on distance and contention.
Generally my ass. Try "always".
Albert, this stuff is all old hat. SGI has done tons of work in this
area and it all turned out to be basically a waste - the Origin SMP
hardware completely smoked any sort of DSM that people did - even when
both were using the same interconnect.
: Don't think I can tell you. (this is coherent memory BTW)
: Of course not.
: No.
: Don't think I can tell you.
I'm sure it was very important for the entire kernel list to know that you
can't tell us. Can you tell us other things you can't tell us?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:18 EST