Novell (NOVL) dirt cheap, good buy? Message Board

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Novell (NOVL) dirt cheap, good buy?

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: scott blomquist who wrote ()	12/18/1999 7:05:00 AM
From: jwright	Read Replies (1) of 42771

Just saw this on PCWeek comparing Windows 2000 server to Windows NT 4.0 server. zdnet.com As a point of comparison I have personally witnessed a single processor server running a beta version of NetWare 5.1 using 60 Windows NT 4.0 workstations as clients achieve almost 1 gigabit/sec throughput (to non techies that would be 1024 Mbit/sec) when running NetBench 6.0. I believe the server was a 450 MHz Xeon with 2 Gigabytes of memory using a Raid 5 disk subsystem. This server was considered state of the art this past summer but now seems kinda slow with the advances made in the last 6 months. It's amazing what a little competition does to advance the technology. It's tough to say just exactly what NetWare 5.1 would achieve without knowing the exact setup Ziff Davis used during their review of Windows 2000 server in the afore mentioned URL since the numbers are impacted by speed of processor(s) in server, memory in server, disk array used in the server, clients used in the test, and the network topology. At home I have server running a beta copy of NetWare 5.1 that has a single processor 500 MHz Pentium III, 512 MB of memory, non RAID disk subsystem (i.e. no concurrent disk writes) achieve 350 Mbit/sec throughput (throughput was still climbing) using 5 Windows NT 4.0 clients (sorry couldn't afford 60 clients) while running NetBench 6.0 on a 100Mbit switched network. If I had a Raid 5 disk subsytem on server I would expect the performance to be near linear or close to 500 Mbit/sec when using same 5 clients. Oh yeah I forget this little nugget. NetWare 5.1 is so fast that you have to reduce the default time the iterations run otherwise the counters that NetBench 6.0 clients use to calculate their throughput ROLL-OVER !!! I believe the default run times are 660 seconds or 11 minutes. We had to reduce them to around 400 seconds. If we didn't the performance curve would look like crap for the first 8 to 12 clients until the server got loaded down enough so that the clients didn't roll over their throughput counters. WOW were so fast we broke the benchmark!!! I sheepishly have to admit it took me a day to figure out this weird behaviour, but was all smiles after I did. Anyway it should be interesting to see what Microsoft, Ziff Davis or Mindcraft report in relation to NetBench when NetWare 5.1 actually ships. I'll probably be in trouble for providing this information on NetWare 5.1 NetBench performance on this board but thems the facts that can be easily verified today by anybody using the NetWare 5.1 that is now in open beta.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next