Comments for blog.richardmcdougall.com http://blog.richardmcdougall.com Commentary on High End Systems and Virtualization Performance Sun, 08 Nov 2009 07:34:50 -0800 http://wordpress.org/?v=2.8.6 hourly 1 Comment on The Solaris Internals Virtual Shell Game by random user http://blog.richardmcdougall.com/?p=80&cpage=1#comment-9433 random user Sun, 08 Nov 2009 07:34:50 +0000 http://blog.richardmcdougall.com/?p=80#comment-9433 Just curious, why vmware as your virtualization layer, instead of Virtualbox? Just curious, why vmware as your virtualization layer, instead of Virtualbox?

]]> Comment on The Solaris Internals Virtual Shell Game by trevor pretty http://blog.richardmcdougall.com/?p=80&cpage=1#comment-8770 trevor pretty Thu, 17 Sep 2009 04:54:49 +0000 http://blog.richardmcdougall.com/?p=80#comment-8770 Richard Nice to see the old Tivo box again. Tivo at long last is coming to New Zealand what's it been 6 years since I first saw Tivo at your place? Ping me off-line. Trevor Richard

Nice to see the old Tivo box again. Tivo at long last is coming to New Zealand what’s it been 6 years since I first saw Tivo at your place?

Ping me off-line.

Trevor

]]> Comment on VMware Performance Tutorial next week at Usenix LISA by satish http://blog.richardmcdougall.com/?p=78&cpage=1#comment-8131 satish Fri, 24 Jul 2009 09:47:46 +0000 http://blog.richardmcdougall.com/?p=78#comment-8131 can you please post me the contents of your tutorial on vmware performance to my mail. can you please post me the contents of your tutorial on vmware performance to my mail.

]]> Comment on The Solaris Internals Virtual Shell Game by Les Armbruster http://blog.richardmcdougall.com/?p=80&cpage=1#comment-8009 Les Armbruster Mon, 13 Jul 2009 04:36:53 +0000 http://blog.richardmcdougall.com/?p=80#comment-8009 How did you make the V-switch for the DMZ's. How did you make the V-switch for the DMZ’s.

]]> Comment on The Solaris Internals Virtual Shell Game by DaveP http://blog.richardmcdougall.com/?p=80&cpage=1#comment-7927 DaveP Fri, 03 Jul 2009 09:05:17 +0000 http://blog.richardmcdougall.com/?p=80#comment-7927 Great write up... it's got me interested in trying something similar and getting rid of some old hardware! What's the case you've used for the Server - looks like a Lian Li but not sure which one? Great write up… it’s got me interested in trying something similar and getting rid of some old hardware!

What’s the case you’ve used for the Server – looks like a Lian Li but not sure which one?

]]> Comment on IO Challenge Update 3: 31,323 IOPS by Alex Low http://blog.richardmcdougall.com/?p=61&cpage=1#comment-7467 Alex Low Sun, 03 May 2009 05:30:36 +0000 http://blog.richardmcdougall.com/?p=61#comment-7467 Hi Richard, I'm an Oracle DBA, one of 4 administering 24x7 Production databases totalling 4 terabytes of data. I've recently come across your writings particularly one pdf titled 'File Systems' for Solaris, and I am highly appreciative of the great technical depth and knowledge you displayed in your deliberation. As an Oracle DBA, I rely heavily on our Solaris Administrators for their contributions and insights. Therefore, my knowledge and experience with Solaris is at best 'basic'. I intend to get your publication from Amazon together with some other definitive books like Jonathan Lewis' on Oracle Optimizer. Would you mind giving me your considered opinion on what is happening with our Solaris I0 I/O sub-system. We have asked the question several times and it seems to have gone 'under the carpet' for some reason. Maybe, it's too challenging I suspect. Oracle's v$filestat and v$tempfile views are telling us that the amount of Oracle related i/os over a given time (after having undergone extensive multiple sql tunings) have gone down significantly. However, using iostat and sar, we don't see a corresponding decrease in demand at the OS level. Even when Oracle is going through a quiet period, the OS is still reporting high kr/s when we expect to see proportionally reduced rates. The question is, what's causing the apparent discrepancy ? We are causing cooked OS ie Veritas FS. So, is it the read-ahead that is giving us more data from the SAN that we need ? What else could we do besides using iostat, var, vmstat ie the usual tools, to find out what might be causing the apparent discrepancy ? Your feedback much appreciated. Thanks. Alex Hi Richard,

I’m an Oracle DBA, one of 4 administering 24×7 Production databases totalling 4 terabytes of data.

I’ve recently come across your writings particularly one pdf titled ‘File Systems’ for Solaris, and I am highly appreciative of the great technical depth and knowledge you displayed in your deliberation.

As an Oracle DBA, I rely heavily on our Solaris Administrators for their contributions and insights. Therefore, my knowledge and experience with Solaris is at best ‘basic’.
I intend to get your publication from Amazon together with some other definitive books like Jonathan Lewis’ on Oracle Optimizer.

Would you mind giving me your considered opinion on what is happening with our Solaris I0 I/O sub-system.

We have asked the question several times and it seems to have gone ‘under the carpet’ for some reason. Maybe, it’s too challenging I suspect.

Oracle’s v$filestat and v$tempfile views are telling us that the amount of Oracle related i/os over a given time (after having undergone extensive multiple sql tunings) have gone down significantly. However, using iostat and sar, we don’t see a corresponding decrease in demand at the OS level. Even when Oracle is going through a quiet period, the OS is still reporting high kr/s when we expect to see proportionally reduced rates.

The question is, what’s causing the apparent discrepancy ? We are causing cooked OS ie Veritas FS. So, is it the read-ahead that is giving us more data from the SAN that we need ?
What else could we do besides using iostat, var, vmstat ie the usual tools, to find out what might be causing the apparent discrepancy ?

Your feedback much appreciated.

Thanks.

Alex

]]> Comment on The Solaris Internals Virtual Shell Game by rmc http://blog.richardmcdougall.com/?p=80&cpage=1#comment-7346 rmc Tue, 21 Apr 2009 06:09:41 +0000 http://blog.richardmcdougall.com/?p=80#comment-7346 I'm using ESX3.5. By default, the UI only shows Raw Device Mapping available for SAN devices, but it is possible to configure SATA and SAS devices in Raw mode using the vmkfstools command at the ESX console. Richard. I’m using ESX3.5. By default, the UI only shows Raw Device Mapping available for SAN devices, but it is possible to configure SATA and SAS devices in Raw mode using the vmkfstools command at the ESX console.

Richard.

]]> Comment on The Solaris Internals Virtual Shell Game by luca http://blog.richardmcdougall.com/?p=80&cpage=1#comment-7240 luca Tue, 07 Apr 2009 12:49:02 +0000 http://blog.richardmcdougall.com/?p=80#comment-7240 Are you using esx or esxi ? Because I'm planning to set up something similar with esxi but I cant configure the internal disks to the VM in raw disk mode. I'm only able to do this with iscsi oder san luns.. maybe there is a difference in esx and esxi.. I'd appreciate your feedback. regards luca Are you using esx or esxi ?

Because I’m planning to set up something similar with esxi but I cant configure the internal disks to the VM in raw disk mode. I’m only able to do this with iscsi oder san luns.. maybe there is a difference in esx and esxi..

I’d appreciate your feedback.

regards
luca

]]> Comment on The Solaris Internals Virtual Shell Game by rmc http://blog.richardmcdougall.com/?p=80&cpage=1#comment-7128 rmc Fri, 27 Mar 2009 22:57:37 +0000 http://blog.richardmcdougall.com/?p=80#comment-7128 # zpool status -v pool: cranky state: ONLINE scrub: none requested config: NAME STATE READ WRITE CKSUM cranky ONLINE 0 0 0 c5t2d0 ONLINE 0 0 0 c5t3d0 ONLINE 0 0 0 c5t4d0 ONLINE 0 0 0 c5t5d0 ONLINE 0 0 0 c5t1d0 ONLINE 0 0 0 logs ONLINE 0 0 0 c5t6d0 ONLINE 0 0 0 # zpool status -v
pool: cranky
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM
cranky ONLINE 0 0 0
c5t2d0 ONLINE 0 0 0
c5t3d0 ONLINE 0 0 0
c5t4d0 ONLINE 0 0 0
c5t5d0 ONLINE 0 0 0
c5t1d0 ONLINE 0 0 0
logs ONLINE 0 0 0
c5t6d0 ONLINE 0 0 0

]]> Comment on The Solaris Internals Virtual Shell Game by rmc http://blog.richardmcdougall.com/?p=80&cpage=1#comment-7127 rmc Fri, 27 Mar 2009 22:55:03 +0000 http://blog.richardmcdougall.com/?p=80#comment-7127 Binny, I didn't need to do any tuning per se, I did however choose a configuration to maximize ZFS's functionallity. Since ZFS handles hot-plug and can do rolling upgrades of larger disks via disk-replace, I wanted to keep the disks fully managed by ZFS. To do this, I used VMware ESX's raw-device-mapping to pass through each disk directly into the VM. The device is still virtualized, but the disks are not, so that ZFS can run exactly the same way as it was on the physical system. Binny,

I didn’t need to do any tuning per se, I did however choose a configuration to maximize ZFS’s functionallity.

Since ZFS handles hot-plug and can do rolling upgrades of larger disks via disk-replace, I wanted to keep the disks fully managed by ZFS. To do this, I used VMware ESX’s raw-device-mapping to pass through each disk directly into the VM. The device is still virtualized, but the disks are not, so that ZFS can run exactly the same way as it was on the physical system.

]]>