Santilli said:So what's the point in getting a dual processor board?
By the way, how do Tyan and Supermico compare with updating their boards for new processors?
In other words, if I buy a Supermicro board, dual for two Xeon 3.06, and, next week, they come out with 3.5ghz models, will supermicro support those with a bios update?
What are the probabilities of a new version of the Xeon, and how high do you think the Xeon chip can be scaled,before Intel changes the pin design for the chip and mobo?
Do the Xeon's have to be matched to work together, like the AMD processors
Santilli said:Stupid Question. It's 3&4, and I've got it.
.Nut said:
Santilli said:Stupid Question. It's 3&4, and I've got it.
SCSI "6" has highest SCSI device priority and is often used for CD/DVD.
SCSI scanners sometimes use "6."
Jan Kivar said:Hmm... I thought that ID7 has the highest priority (assigned to the HBA), and then it goes from ID0 to ID6, ID6 having the lowest priority.
PLEASE correct me if I'm wrong.
Platform said:No need to mention "7" in the PC world, since we know a SCSI peripheral isn't normally setup for using "7" (older SGI boxes reserve "4" for its SCSI controller).
Once SCSI-2 was introduced, highest to lowest in SCSI priority became:
7, 6, 5, 4, 3, 2, 1, 0, 15, 14, 13, 12, 11, 10, 9, 8.
Another consideration is that if you have any devices that absolutely cannot tolerate delays in receiving their stream of data--such as a CD recording drive or a video encoder--they should be given top priority on the bus. Many people also like to make the host adapter the highest-priority device on the bus, which is why host adapters will often default to a SCSI ID of 7. It should be noted that some older host adapters can be finicky about device IDs. Some will only boot a hard disk if it is set to device ID 0. (This is inflexible and has been basically done away with in newer hardware.)
Handruin said:That's very interesting... I found more (of the same) info here.
Jan, the link I posted says:
Another consideration is that if you have any devices that absolutely cannot tolerate delays in receiving their stream of data--such as a CD recording drive or a video encoder--they should be given top priority on the bus. Many people also like to make the host adapter the highest-priority device on the bus, which is why host adapters will often default to a SCSI ID of 7. It should be noted that some older host adapters can be finicky about device IDs. Some will only boot a hard disk if it is set to device ID 0. (This is inflexible and has been basically done away with in newer hardware.)
mubs said:Anytime the issue of a dual-proc system comes up, there seems to be much gnashing of teeth...
...Even if every program one runs is SMP- un aware, is there not a benefit to having a second CPU?
...A CPU eating program like Santilli's possessed Photoshop can hog one processor,...
...and the OS and all other progs like mail client, browser, etc. can time-slice on the other. Will this not make for a more responsive system overall? I would even say that in a situation like this, there is a benefit to having a program not being multi-threaded because then it can't hog the second processor as well.
...So long as the OS can use dual CPUs, I think there is still a benefit. Apparently, Opterons scale 80-90% compared to Xeons which scale at only 20-30% (saw this somewhere - Inquirer?)...
Well, it is possible to have highly efficient scalability with SMP. Dunno what Intel is doing but in AS/400-land, we had a 2-way machine with a performance rating of 1050 CPWs (IBM variant on TPC-C). Upping to a 4-way brought the perf rating to 2000. About 90% scalability.Corvair said:...So long as the OS can use dual CPUs, I think there is still a benefit. Apparently, Opterons scale 80-90% compared to Xeons which scale at only 20-30% (saw this somewhere - Inquirer?)...
I'm not sure I would believe anything they say in the Inquirer, much less the National Enquirer. Isn't the Inquirer some sort of bad joke???
I don't believe they are; I'll have to look further. A few years ago when IBM was buying SEQUENT, I was at a user conference. The IBMers there were hyped because IBM was said to be just developing their NUMA tech and hoped for a good jump start from teh SEQUENT tech. So I'll say early boxes were not but I can't yet say about current machines.Mercutio said:Are AS400s NUMA systems?
Fushigi said:I don't believe they are; I'll have to look further.Mercutio said:Are AS400s NUMA systems?
Santilli said:Which handles SMP better, 2000 or XP?
Mercutio said:I'd suggest something with a name that ends in "Server", actually, since those products are a little better tuned for the large amounts of RAM that tend to go hand-in-hand with MP machines.Santilli said:Which handles SMP better, 2000 or XP?
Santilli said:Are the DVD RW format a viable solution for large storage?
The firewire writer I've been using is really cool, allowing a regular CD-R to be rewritten a number of times, for document backups. Does the DVD RW format work the same way, and is it fast enough?
Thanks
gs
Santilli said:What does Prescott bring to the table that would make it worth waiting for?
Dual cores on the same chip?
Do Xeons have the same features?
No, they run better.Santilli said:Does windows run as well on the Opteron as it does on the xeons?
Buck said:Boy, putting together a dual Opteron 240 system with the S2885 board and a couple gigs of Reg./ECC PC2700 memory really gets expensive. Add a really good power supply and nice case, and you easily hit $2,500.00.
Santilli said:WOW!!! :eekers:
:bounce:
It's amazing how slow the clock speed is, yet how fast the results are with the Opteron.
Why?
gs