SUMMARY: 9Gb disks run slower than 4Gb???RE:

From: <franx_at_oas.telstra.com.au>
Date: Fri, 19 Dec 1997 12:32:58 +1100

L.O. (again!),

Thanx to:

alan_at_nabeth.cxo
Andy Walden
Dr. Tom Blinn
Guy Dallaire
Max Pabia
Paul Grant
Peter.Braack
Tom Webster

ESPECIALLY to Peter who suggested turning the writeback cache
on at the HSZ50 controller:

        STOAT> show d300
         
            LUN Uses
        --------------------------------------------------------------
         
          D300 R4
                Switches:
                  RUN NOWRITE_PROTECT READ_CACHE
                  NOWRITEBACK_CACHE
                  MAXIMUM_CACHED_TRANSFER_SIZE = 32
                State:
                  ONLINE to the other controller
                  PREFERRED_PATH = OTHER_CONTROLLER
                Size: 53294505 blocks
        STOAT> set d300 writeback_cache
         
        Warning 1000: It is recommended that you read the controller product
                      documentation to understand the significance of enabling
                      WRITEBACK_CACHE particularly for RAID Arrays
        STOAT> show d300
         
            LUN Uses
        --------------------------------------------------------------
         
          D300 R4
                Switches:
                  RUN NOWRITE_PROTECT READ_CACHE
                  WRITEBACK_CACHE
                  MAXIMUM_CACHED_TRANSFER_SIZE = 32
                State:
                  ONLINE to the other controller
                  PREFERRED_PATH = OTHER_CONTROLLER
                Size: 53294505 blocks
        STOAT> ^D
        %108 toad[alpha]/u01/home/franx>

(A bit emabarrassing on my part, but a commendable guess on Peters)

ALSO

from Paul from Scotland (I must get back there 1 day):
 
> the throughput of two 4.3 gbyte disks is 2 * 140 i/o's per second
> (depending on type of usage).

> the throughput of one 9.1 gbyte disk is 1 * 140 i/o's per second (ditto)
 
> Therefore there will be loss of performance
> as you have half the io potential of the raidsets.....

AND

the final word from Max:

> The 9G 7200rpm drives are slower than the 4G 7200 rpm
> drives. Dec will be coming out with 10000rpm 9G drives in a few
> months to make up the difference.
 
> This argument with bigger drives being slower than smaller
> drives have been going on forever. When they first came out with
> 4G 5400rpm drives, the problem then was that the 2G 5400 rpm
> drives were faster than these 4G drives. Then 4G 7200rpm
> drives came out to speed them up.
 
> Don't buy too many 9G drives yet, and wait for the 10000 rpm
> 9G drives to come out.
 
Once again, a big "THANKYOU" to everyone who responded; everyone
seemed to have interesting & plausible suggestions on this one.

HAVE A COOL YULE & A GREAT '98

TNX1.0E6, fcG.
-- 
Frank Gallacher, Sys. ANALyst/Programmer,(aka.frAnXoSAURus reX)
Systems D&I, Operator Assisted Services, Telstra (TELecom auSTRAlia).
 ___________
(_'-I---I-`_)		THE POLITICALLY CORRECT VIRUS:
  | 1 2 3 |
  | 4 5 6 |		Never calls itself a "virus", but instead
  | 7 8 9 |		refers to itself as an "electronic microorganism."
  | * 0 # |
  +-------+
phone: (03) 969 34034		Int.: +61 3 969 34034
smtp:  franx_at_oas.telstra.com.au, FGallach_at_VENTNOAS.telecom.com.au
X.400: (C:AU,A:TELEMEMO,P:telecom012,O:TELECOM,OU:VENTNOAS,SN:FGallach)
<<<MSG ENDS>>>
Received on Fri Dec 19 1997 - 02:31:52 NZDT

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:37 NZDT