#Huh? 1200 diskspeed, NOT
15-May-94 03:10:22
Sb: #145720-#Huh? 1200 diskspeed, NOT
Fm: Paul Idol 71024,3245
To: Chris Lester 73553,1670
Well, now I'm confused. I just did the whole thing over again and this is
what I got -this- time:
————————————————————————-
9.RAM:> type DiskSpeed.ResultsCache
DiskSpeed 3.0 – Copyright (c) 1989,90 by MKSoft Development
Device: work:
Test Intensity: High Performance Stress: CPU Contention & DMA Contention
54 Files/s Create
202 Files/s Open/Close
1422 Files/s Scan
286 Files/s Delete
734 Seek/Read
Buffer Size 512 4096 32768 262144
——— ——— ——— ——— ———
Bytes/s Create 320519 1675104 3303820 3962146
Bytes/s Write 357199 1997659 4277855 5052902
Bytes/s Read 366715 2049125 4194304 5413824
9.RAM:> type DiskSpeed.ResultsnoCache
DiskSpeed 3.0 – Copyright (c) 1989,90 by MKSoft Development
Device: work:
Test Intensity: High Performance Stress: CPU Contention & DMA Contention
5 Files/s Create
30 Files/s Open/Close
739 Files/s Scan
131 Files/s Delete
707 Seek/Read
Buffer Size 512 4096 32768 262144
——— ——— ——— ——— ———
Bytes/s Create 27585 94636 126829 135830
Bytes/s Write 29143 112663 163021 177889
Bytes/s Read 164987 311801 303231 314572
————————————————————————-
So color me confused. The PowerCache-less results this time were higher
than the cached AND cache-less results last time. What changed? Nothing.
I left the machine on, didn't quit any applications, and went out to see a
movie. When I came back I got some mail here and then ran the test while
I took a shower. I am totally baffled.
In addition, I don't know how to translate the results to real-world
performance. I just loaded Final Writer and it STILL takes about 7 beats
to load. So no change there. And while I don't know that all reads and
writes are 512bytes, I doubt most of them are 262144bytes long either. So
how do things work in mixed situations? ACK!
Paul