- Community Home
- >
- Servers and Operating Systems
- >
- Legacy
- >
- HPE 9000 and HPE e3000 Servers
- >
- Re: SuperDome Repartitioning
HPE 9000 and HPE e3000 Servers
1752689
Members
5358
Online
108789
Solutions
Forums
Categories
Company
Local Language
юдл
back
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
юдл
back
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
Blogs
Information
Community
Resources
Community Language
Language
Forums
Blogs
Topic Options
- 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
тАО09-04-2003 06:58 PM
тАО09-04-2003 06:58 PM
SuperDome Repartitioning
I have a 32 way Superdome and it is currently split into 4 nPars as follows:
Partition 0 = Cells: 0 & 2 (0 is core)
Partition 1 = Cells: 1 & 3 (1 is core)
Partition 2 = Cells: 4 & 6 (4 is core)
Partition 3 = Cells: 5
There is 1 cell board that had not been assigned to a partition until we determined which part of the application and/or database would need the extra horse-power. We have determined that partition 0 will require the extra resources. The problem is that according to the Partitions guide the supported configuration needs to look like:
Partition 0 = Cells: 0, 1 & 2 (0 is core)
Partition 1 = Cells: 5 & 7
Partition 2 = Cells: 4 & 6
Partition 3 = Cells: 3
Its no problem moving the drives, etc. around to get this configuration, but there is a problem with the fact that currently cell #1 is attached to a core I/O and cell #3 is not. I know there is some internal cabling that needs to be moved, but I don't know what else needs to be done. I couldn't find this in any manuals.
Our environment is very dynamic and I know that management would not be happy if we have to call HP Services out every time we need to reconfigure one of our superdomes.
At this point my options appear to be:
1. Get HP Services to come out and rework the cabinet
2. Find a manual or some instructions on how to rewire the core I/O <-> Cell.
3. Just add cell #7 into partition 0 as it is. I don't know what kind of performance hit this would be (if any), but I know it is not a supported configuration.
Any suggestions would be greatly appreciated.
Thanks,
David
Partition 0 = Cells: 0 & 2 (0 is core)
Partition 1 = Cells: 1 & 3 (1 is core)
Partition 2 = Cells: 4 & 6 (4 is core)
Partition 3 = Cells: 5
There is 1 cell board that had not been assigned to a partition until we determined which part of the application and/or database would need the extra horse-power. We have determined that partition 0 will require the extra resources. The problem is that according to the Partitions guide the supported configuration needs to look like:
Partition 0 = Cells: 0, 1 & 2 (0 is core)
Partition 1 = Cells: 5 & 7
Partition 2 = Cells: 4 & 6
Partition 3 = Cells: 3
Its no problem moving the drives, etc. around to get this configuration, but there is a problem with the fact that currently cell #1 is attached to a core I/O and cell #3 is not. I know there is some internal cabling that needs to be moved, but I don't know what else needs to be done. I couldn't find this in any manuals.
Our environment is very dynamic and I know that management would not be happy if we have to call HP Services out every time we need to reconfigure one of our superdomes.
At this point my options appear to be:
1. Get HP Services to come out and rework the cabinet
2. Find a manual or some instructions on how to rewire the core I/O <-> Cell.
3. Just add cell #7 into partition 0 as it is. I don't know what kind of performance hit this would be (if any), but I know it is not a supported configuration.
Any suggestions would be greatly appreciated.
Thanks,
David
3 REPLIES 3
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО09-05-2003 06:39 AM
тАО09-05-2003 06:39 AM
Re: SuperDome Repartitioning
I have come up with one other option. It is a variation of #3:
4.
Partition 0 = Cells: 0 & 2 & 3
Partition 1 = Cells: 1
Partition 2 = Cells: 4 & 6
Partition 3 = Cells: 5 & 7
This way partitions 1, 2, and 3 are supported configurations. Partition 0 will not be, but at least all four cells are in the same quad and shouldn't run into the performance issue with adding cell 7 into an nPar with cells 0 and 2.
Any thoughts?
David
4.
Partition 0 = Cells: 0 & 2 & 3
Partition 1 = Cells: 1
Partition 2 = Cells: 4 & 6
Partition 3 = Cells: 5 & 7
This way partitions 1, 2, and 3 are supported configurations. Partition 0 will not be, but at least all four cells are in the same quad and shouldn't run into the performance issue with adding cell 7 into an nPar with cells 0 and 2.
Any thoughts?
David
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО09-08-2003 01:27 AM
тАО09-08-2003 01:27 AM
Re: SuperDome Repartitioning
Hi David,
I think that there is no need for moving internal cabling inside Superdome. Configuration Guide for Superdome tells that "resources may be removed from one partition and added to another without having to physically manipulate the hardware just by using commands that are part of the System Management interface".
BR,
OK
I think that there is no need for moving internal cabling inside Superdome. Configuration Guide for Superdome tells that "resources may be removed from one partition and added to another without having to physically manipulate the hardware just by using commands that are part of the System Management interface".
BR,
OK
No RISC, no fun
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО09-08-2003 05:21 AM
тАО09-08-2003 05:21 AM
Re: SuperDome Repartitioning
Thanks Oleg,
Yes, technically you can reassign resources as required, but HP only supports a limited number of configurations. As I listed above, I cannot put together the 4 nPar configurations I want in an HP supported way without recabling.
I reworked the system this weekend and went with my last suggestion:
Partition 0 = Cells: 0 & 2 & 3
Partition 1 = Cells: 1
Partition 2 = Cells: 4 & 6
Partition 3 = Cells: 5 & 7
Partitions 1, 2, and 3 are in supported configurations. Partition 0 will not be, but I kept all three cells in the same quad for performance.
For future reference: if you are setting up a 32way superdome with several 2 cell nPars, make sure you design it so cell 0 and cell 3 are attached to core I/O as are cells 4 and 7. This way you can have supported 2 cell nPars AND you can easily make a 3 cell nPar in a supported configuration. This applies to 16 say superdomes as well.
Yes, technically you can reassign resources as required, but HP only supports a limited number of configurations. As I listed above, I cannot put together the 4 nPar configurations I want in an HP supported way without recabling.
I reworked the system this weekend and went with my last suggestion:
Partition 0 = Cells: 0 & 2 & 3
Partition 1 = Cells: 1
Partition 2 = Cells: 4 & 6
Partition 3 = Cells: 5 & 7
Partitions 1, 2, and 3 are in supported configurations. Partition 0 will not be, but I kept all three cells in the same quad for performance.
For future reference: if you are setting up a 32way superdome with several 2 cell nPars, make sure you design it so cell 0 and cell 3 are attached to core I/O as are cells 4 and 7. This way you can have supported 2 cell nPars AND you can easily make a 3 cell nPar in a supported configuration. This applies to 16 say superdomes as well.
The opinions expressed above are the personal opinions of the authors, not of Hewlett Packard Enterprise. By using this site, you accept the Terms of Use and Rules of Participation.
News and Events
Support
© Copyright 2024 Hewlett Packard Enterprise Development LP