A computer components & hardware forum. HardwareBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » HardwareBanter forum » General Hardware & Peripherals » Storage & Hardrives
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

incredibly overloaded disk subsystem



 
 
Thread Tools Display Modes
  #1  
Old December 10th 03, 01:53 AM
aDF
external usenet poster
 
Posts: n/a
Default incredibly overloaded disk subsystem

per MS, disk queue length values as measured by PerfMon on Win2K should be
very low (1-2), or you're experiencing I/O issues.

well, we have an app powered by an Oracle database that's constantly getting
hammered. We routinely measure queue lengths in the hundreds on up to the
6,000-7,000 range. As you might imagine, our app works very poorly & we
have lots & lots of problems. I'm convinced it's the database I/O that
kills us, to the point that it's what makes our com servers that talk to it
fail, things like that.... like requests are just getting blocked, that sort
of thing.

is this reasonable? has anyone ever done any sort of study that indicates
odds of failure as you push the hardware well beyond its limits? We have
two of Dell's high-end controllers powering this (PERC3/DC pushing two RAID1
sets for archive logs and another PERC3/DC powering a RAID10 set for
database). The OS (Win2K AS SP4) is on another separate RAID adapter. I'm
convinced there is nothing we can do to help ourselves until we remove this
bottleneck, & i also believe we are putting ourselves at serious risk of
catastrophic failure with our current system.

i appreciate any thoughts.

thanks


  #2  
Old December 10th 03, 10:57 AM
Tarjei T. Jensen
external usenet poster
 
Posts: n/a
Default

"aDF" wrote:
well, we have an app powered by an Oracle database that's constantly

getting
hammered. We routinely measure queue lengths in the hundreds on up to the
6,000-7,000 range. As you might imagine, our app works very poorly & we
have lots & lots of problems. I'm convinced it's the database I/O that
kills us, to the point that it's what makes our com servers that talk to

it
fail, things like that.... like requests are just getting blocked, that

sort
of thing.


First, make sure that Oracle works smoothly. A misconfigured Oracle database
can generate a LOT of I/O.

Go to http://oracle.oreilly.com/ Choose "Optimizing Oracle Performance".
Read the sample chapter. Buy the book and read it or contact a reputable the
Oracle tuning company (eg. the company the author works for). Tune database.

When you know that Oracle is tuned, then you should consider everything
else.

greetings,



 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
CUV4X, BIOS, and MAx Disk size? JoeM Asus Motherboards 2 December 22nd 03 04:22 AM
Boot problem from copied HD - Partitionmagic 7.0, Win 2000 DRIZAII Homebuilt PC's 8 December 10th 03 09:28 PM
48-bit LBA Support - HDs Over 137GB - Advice? Rob Jones Asus Motherboards 16 November 30th 03 12:20 PM
Problem booting from SATA disk with GA-8KNXP motherboard (like many others) Eric Janvier Gigabyte Motherboards 9 November 22nd 03 01:56 AM
Lost Hard Disk Robert Brown Homebuilt PC's 2 October 9th 03 01:13 AM


All times are GMT +1. The time now is 08:31 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 HardwareBanter.
The comments are property of their posters.