- Community Home
- >
- Servers and Operating Systems
- >
- HPE ProLiant
- >
- ProLiant Servers (ML,DL,SL)
- >
- Re: ProLiant DL380 G3 I/O Bottleneck
Categories
Company
Local Language
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Community
Resources
Forums
Blogs
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 12:57 AM
тАО06-22-2004 12:57 AM
Re: ProLiant DL380 G3 I/O Bottleneck
Do you know if your disks are fragmented? you could run a fragmentation anaylsis from within Windows and let us know what report you get.
Do you have a large number of small files on the NTFS volume? What about CPU and memory utilisation on the box? I know SQL can chew away massive amounts of memory, have you actually gone through SQL tweaking?
please reply to the above and see how we go...
hope it helps and dont forget to assign points:)
regards,
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 11:19 AM
тАО06-22-2004 11:19 AM
Re: ProLiant DL380 G3 I/O Bottleneck
I have some file fragmentation but nothing too serious.
I have been through the tweaking of SQL server also.
We are not experiencing any paging so I have assumed memory is not an issue.
Would setup of cabling etc be an issue for through put? Is there a way of determining i/o transfer rates?
Andrew
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 01:50 PM
тАО06-22-2004 01:50 PM
Re: ProLiant DL380 G3 I/O Bottleneck
The 6402 also has 128MB read/write cache.
Is anyone able/willing to try and determine a performance increase from the 5i to a 6402 controller? :-)
Andrew
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 02:19 PM
тАО06-22-2004 02:19 PM
Re: ProLiant DL380 G3 I/O Bottleneck
you could use the good old IOmeter tool to isolate I/O performance issues.
Below is the link for it:
http://sourceforge.net/projects/iometer/
hope it helps and dont forget to assign points:)
regards,
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 05:52 PM
тАО06-22-2004 05:52 PM
Re: ProLiant DL380 G3 I/O Bottleneck
The BBWC is an installable option on the DL380 G3, so if you don't get the option to turn on the relevant cacheing it more than likely isn't installed.
You can find out for certain by taking the lid off, and looking for the battery holder which slots in to the front right hand side of the server.
The picture here:
http://h18004.www1.hp.com/products/servers/proliantstorage/arraycontrollers/smartarray5iplus/index.html
shows the battery module.
Regards,
Rob
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 05:57 PM
тАО06-22-2004 05:57 PM
Re: ProLiant DL380 G3 I/O Bottleneck
there will be sure performance increase if you migrate to SA 6402, bacause, not only you will have optimum speed handelling ( UW320) for youe HDDs, but you will be able to set read /write chaching as desired.one thing I want to know is , what is the array configuration in your system? ( I'm sorry , you have written RAID 1, but raid 1 can only have 2 disks, so, how you are using the third one?).
regards,
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 06:52 PM
тАО06-22-2004 06:52 PM
Re: ProLiant DL380 G3 I/O Bottleneck
I would recommend you take a look at the following link:
http://h18000.www1.hp.com/products/quickspecs/11063_na/11063_na.HTML
regards,
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-22-2004 10:04 PM
тАО06-22-2004 10:04 PM
Re: ProLiant DL380 G3 I/O Bottleneck
Saket - Have just downloaded IOmeter, will give it a shot and get back to you all with the results.
Rob - Thanks for that info, now deciding if we are better off just upgrading to the 6402 controller instead of getting a BBWC enabler.
kcpant - It would be nice to know what kind of performance increase we are likely to see, has anyone upgraded from a 5i to a 6402? If so, what was the difference?
Our RAID is a RAID 1 with an online spare. A more expensive option, however we run a critical application and redundancy is a big priority.
I may make a post in another forum to see if anyone has upgraded from a 5i to a 6402 controller to try and measure the improvement. Please continue this thread however.
Andrew
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-23-2004 01:43 AM
тАО06-23-2004 01:43 AM
Re: ProLiant DL380 G3 I/O Bottleneck
DL380g3 uses Simplex setup as standard --- Just want to check whether you have already set it to Duplex configuration for you drive cage?
SA6402/4 is such high-end controller it will certainly give you performance boost... however, people will use that mostly to connect to things like StorageArrays30DB... But I don't think it is good value for money, as Duplexing the 5i should already give you max 320MB/s (160MB/s per channel).
However, if you can easily afford such controller (hey, SA641 could be ideal too), I would also recommend you to buy another hard drive (to make it up to 2 sets of RAID1). Then, set the DL380g3 to PCI Duplex so that you'd get 480MB/s max throughput (160MB/s in slot 0,1 controlled by the 5i for OS usage, and 320MB/s shared on slot 2,3,4,5 controlled by SA6402/SA641 for MSSQL usage)... Of course, you can use PCI Simplex and let the SA6402 to handle all 6 slots at shared max 320MB/s --- that will depend on your environment.
Hope this helps.
Cheers,
Gus
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-23-2004 02:29 AM
тАО06-23-2004 02:29 AM
Re: ProLiant DL380 G3 I/O Bottleneck
All of the suggestions about upgrading controllers and RAID sets is good, but I am convinced that the single most important thing you could do is to add enough drives to physically separate your OS, data, and transaction logs.
Ideally you would have 3 arrays, each with their own single disk volume, and each of them set up as a 2 drive RAID 1. Another option wuold be to combine the OS and data onto a single RAID 1 array (2 drives) with 2 volumes.
However, the MOST important thing for database performance - especially with heavy writes - is to dedicate a physical array (2 drives in RAID 1) with a single volume.
Think about it this way... the ACU allows you to set up multiple volumes within an array of physical drives. This is nice for convenience of assigning LUNs, but it can be a performance killer for databases.
OS and data volumes are characterized by random read/write I/O. So combining them (with 2 volumes) on a single physical 2 drive array is no big deal - it's going to do random I/O anyways.
But the transaction logs are characterized by sequential writes - that's it. By combining the volume containing the logs on the same physical array as the random I/O of the OS and data, you really reduce the write performance of the transaction log (which is already very busy because you are in a write intensive environment). It all comes down to head movement in the physical drives themselves. With random I/O the heads are moving all over the place, back and forth. With sequential writes, the heads IDEALLY only move in teeny little increments to do the next sequential write. By combining them, you have 2 opposite performance requirements - the large amount of head movement back and forth across the disk platters is in direct competition with the teeny little movements of the sequential writes.
As a side note, having a Smart Controller with a relatively large BBW Cache will also do wonders for performance, but this will have less of an effect than separating the log files (you should really do both!)
Check out these white papers for more info:
http://h71019.www7.hp.com/ActiveAnswers/Render/1,1027,5108-6-100-225-1,00.htm
http://h71019.www7.hp.com/ActiveAnswers/Render/1,1027,5264-6-100-225-1,00.htm
http://h71019.www7.hp.com/ActiveAnswers/Render/1,1027,5265-6-100-225-1,00.htm
Thanks,
Doug