| TIP: Click on subject to list as thread! | ANSI |
| echo: | |
|---|---|
| to: | |
| from: | |
| date: | |
| subject: | HPFS386 performance for programmers |
CS> I still don't agree with your assessment on this CS> for computers with lots CS> of RAM (i.e., 32MB RAM or more) and fast CPU chips CS> (Pentium or faster). For instance, a Pentium with CS> 64MB RAM being used for heavy disk I/O can probably CS> benefit from the larger than 2MB cache size available CS> with HPFS386. PF> You can always design a test to skew the results in PF> the direction you want. That's true. PF> My own tests statisfy me that in all but a few PF> isolated (ie: dedicated task PC's) real world PF> situations OS/2 always does better with more RAM for PF> itself, than stolen by a vdisk or disk cache. PF> I have 32mb of ram, run hpfs386, and only use a 2mb cache. Using a PF> larger cache will cause more swapping (which is not PF> cached at all; a catch-22 by necessity). I've got 64MB RAM on this P54C/100 system with a fast SCSI-2 hard drive (Seagate ST12550N "Barracuda") and was using a 4MB HPFS386 cache. I just tried an experiment of increasing it to 8MB. My project rebuild time dropped from about 84 minutes to 53 minutes with the larger cache. The CACHE386 /STATS also showed an increase in the cache read hit rate from 80% to 95%. This wasn't a completely "controlled" experiment in that the source code changed slightly between builds and I might not have all the same software running, so I'll have to try this again under more controlled circumstances to be sure about the level of improvement. But it does seem that the 8MB cache is getting a much higher read hit rate and does improve performance on this type of task. PF> I think people will see a much larger change in PF> performance playing with the write-back times and the PF> /CRECL value than they will with a >2mb disk cache. If you've 16MB RAM or less, I think that is probably almost always true. But once you're up in the 32MB+ range, that may not be the case across the board. I will say, however, that for VisualAge C++ 3.0 on a 32MB RAM system, you can't afford much more than a 2MB cache anyway. VAC++ 3.0 is a major memory pig, even without using Visual Builder. Thankfully RAM prices have been dropping lately. I can see that within a couple of years it might sound so unusual to be using computers with 128MB RAM or more. --- Maximus/2 2.02* Origin: OS/2 Connection {at} Mira Mesa, CA (1:202/354) SEEN-BY: 50/99 270/101 620/243 711/401 409 410 413 430 808 809 934 955 SEEN-BY: 712/407 515 517 628 713/888 800/1 7877/2809 @PATH: 202/354 300 777 3615/50 396/1 270/101 712/515 711/808 809 934 |
|
| SOURCE: echomail via fidonet.ozzmosis.com | |
Email questions or comments to sysop@ipingthereforeiam.com
All parts of this website painstakingly hand-crafted in the U.S.A.!
IPTIA BBS/MUD/Terminal/Game Server List, © 2025 IPTIA Consulting™.