RAID1 vs RAID10?
in [Microsoft Exchange]
|
Prev: 5.7.1 smtp;550 5.7.1 Requested action not taken: message refused
Next: How to enable Intelligent Message Filter auto update?
From: John Fullbright on 14 Jan 2006 12:34 Yes, that's Nicole Allen's blog. The only thing I don't like about is the IOPS per spindle numbers have no time associated with them. Yes, you can load a 10K spindle to 130 IOPS, however your IO times will increase as you load the spindle up. At 130 IOPS/10K spindle, we're talking 80 to 100ms IOs which are unacceptable for Exchange. "Paul Hutchings" <paul(a)spamcop.net> wrote in message news:paul-262D4D.12345214012006(a)msnews.microsoft.com... > In article <#pIEICKGGHA.1192(a)TK2MSFTNGP11.phx.gbl>, > "John Fullbright" <Fullbrij(a)comcast.com> wrote: > >> Ok, you left out the pessimism. I actually do storage design for Exchange >> for a storage vendor and enjoy it when competitors string the customer >> along >> like that. When the customer gets the sticker shock 12 to 18 months down >> the line, the competative takeout is a no brainer. > > The HP sizer was a little more realistic I think - it suggested more or > less what I had in mind (it said less memory and slower CPUs which I > shall ignore), the main difference being an external enclosure because > of the number of spindles with dedicated disks for the transaction logs, > and spares. > > I could have a dedicated mirror for the logs if I sacrificed the > hot-spare and the spindle for the NT backup , BUT it would be on the > same RAID controller channel as the databases, so I'm not sure if I'd be > any better off than having the OS/logs on the same spindles but on a > different RAID controller channel. > > So far as measuring what level of disk performance I need, from what > I've read this seems useful (from > http://blogs.technet.com/exchange/archive/0001/01/01/240868.aspx) > > Measure the physical disk\disk transfers per second for all databases > for between 20 minutes to 2 hours during your most active time (for > example, this is from 9-11 AM on a Monday here at Microsoft). During > this time, also measure the number of active users (MSExchangeIS\Active > User Count). Take an average of these counters. Sum the disk > transfers/sec for each database, divide the first number by the second > andS Voila! You have just calculated the number of IOPS per user. > > Paul > -- > paul(a)spamcop.net
From: Paul Hutchings on 16 Jan 2006 08:04 "John Fullbright" <Fullbrij(a)comcast.com> wrote in news:#7eB4BTGGHA.984(a)tk2msftngp13.phx.gbl: > Yes, that's Nicole Allen's blog. The only thing I don't like about is > the IOPS per spindle numbers have no time associated with them. Yes, > you can load a 10K spindle to 130 IOPS, however your IO times will > increase as you load the spindle up. At 130 IOPS/10K spindle, we're > talking 80 to 100ms IOs which are unacceptable for Exchange. I've been logging the Disk Transfers/Sec every second against the database drive for a couple of hours now as it's an average working day. I'm shocked. Minimum 0, Maximum 861, average of 45.721. With 400 users maximum this gives 0.11 IOPS per user. Seems low but the numbers can't lie can they? Paul
From: John Fullbright on 16 Jan 2006 20:23
Read the paper "Optimizing Storage for Exchange 2003". Use the peak, not the average. In fact, "Always design your system to allow 20 percent more utilization than you expect for peaks. This allows the storage and processors to handle spikes during peak periods." If you use the average value, you will experience performance problems as evidenced by your peak value of 861. With a peak of 861 and 400 users, your IOPS value is 2.305 without adding the 20% as the paper recommends. "Paul Hutchings" <paul(a)spamcop.net> wrote in message news:Xns974D84ECEF8D5paulspamcopnet(a)207.46.248.16... > "John Fullbright" <Fullbrij(a)comcast.com> wrote in > news:#7eB4BTGGHA.984(a)tk2msftngp13.phx.gbl: > >> Yes, that's Nicole Allen's blog. The only thing I don't like about is >> the IOPS per spindle numbers have no time associated with them. Yes, >> you can load a 10K spindle to 130 IOPS, however your IO times will >> increase as you load the spindle up. At 130 IOPS/10K spindle, we're >> talking 80 to 100ms IOs which are unacceptable for Exchange. > > I've been logging the Disk Transfers/Sec every second against the database > drive for a couple of hours now as it's an average working day. > > I'm shocked. Minimum 0, Maximum 861, average of 45.721. > > With 400 users maximum this gives 0.11 IOPS per user. > > Seems low but the numbers can't lie can they? > > Paul |