Re: disk subsystem performance question

2002-04-11 Thread Gaja Krishna Vaidyanatha
as additional experiences or feedback > with Hitachi or EMC they > > would like to share or comments (agree/disagree) > with my thoughts, I'd love > > to hear them. I'm open for learning! > > > > Thanks, > > > > John Dailey > > Oracle DBA &g

Re: disk subsystem performance question

2002-04-11 Thread Igor Neyman
Battery backup on the controller should help in this case. If there is none, "write back" mode should be used (instead of "write through"), loosing advantages of on-board caching (still helps with the "reads"). Igor Neyman, OCP DBA [EMAIL PROTECTED] - Original Message - To: "Multiple r

Re: disk subsystem performance question

2002-04-11 Thread CHRIS FARMER
Marie, there is a thread going on this Oracle list about disk subsystem speed, I am going to forward a couple of the responses to you. I think that it is reiterating that we need to have raid 0+1 in our PROD environment. >>> [EMAIL PROTECTED] 04/10/02 01:06PM >>> James Howerton wrote: > > J

RE: disk subsystem performance question

2002-04-11 Thread paquette stephane
; with my thoughts, I'd love > to hear them. I'm open for learning! > > Thanks, > > John Dailey > Oracle DBA > ING Americas - Application Services > Atlanta, GA > > > > > > > "Don > >

RE: disk subsystem performance question

2002-04-10 Thread Deshpande, Kirti
I'm open for learning! Thanks, John Dailey Oracle DBA ING Americas - Application Services Atlanta, GA "Don Granaman"To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>

Re: disk subsystem performance question

2002-04-10 Thread Charlie Mengler
> "Don > Granaman" To: Multiple recipients of list >ORACLE-L <[EMAIL PROTECTED]> > .net>Subject: Re: disk subsystem performance >qu

Re: disk subsystem performance question

2002-04-10 Thread John . Dailey
Granaman"To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]> Subject: Re: disk subsystem performance question

Re: disk subsystem performance question

2002-04-10 Thread Peter Barnett
Much of the supposed 'speed' may come from onboard caching on the controller. There is the minor risk that a crash could come after Oracle commits the data and before it is actually written to disk. --- [EMAIL PROTECTED] wrote: > Hi all, > > We are running both a Hitachi 7700E and a 9960 disk >

RE: disk subsystem performance question

2002-04-10 Thread Deshpande, Kirti
John, We use 7700E. Most all EMC is going out and being replaced by HDS RAID-5. (What it costs is more important that what it does). Everything we have is on RAID-5 these days. HDS Techies and our Capacity Planners (who do the disk assignments to Servers) told me the following: The 18GB drives a

Re: disk subsystem performance question

2002-04-10 Thread Charlie Mengler
James Howerton wrote: > > John, > > We have the Hitachi 5800 series with RAID 5. The sales guys also said > their system is > "s fast we need not worry about such minor details". Don't believe > them!!! Write speed is "SLOW". After we added bare drives for redo log > files, archive logs, &

Re: disk subsystem performance question

2002-04-10 Thread Don Granaman
Short answer - NO! Nobody's disk subsystem is so fast that no intelligence is required in the layout. This is common vendor blather and one of the most popular myths. I have been hearing it for at least six years - and it still isn't true. Layout still makes a huge difference. RAID levels sti

Re: disk subsystem performance question

2002-04-10 Thread James Howerton
John, We have the Hitachi 5800 series with RAID 5. The sales guys also said their system is "s fast we need not worry about such minor details". Don't believe them!!! Write speed is "SLOW". After we added bare drives for redo log files, archive logs, & conrtol files it made a dramatic differe

RE: disk subsystem performance question

2002-04-10 Thread DENNIS WILLIAMS
John - I haven't experienced the Hitachi disks. To me the key question is how much battery-backed cache the subsystem has. I believe there tends to be a gap in knowledge between DBAs and the disk people. The two don't always understand each other or speak the same language. The Oracle books tend t