- Community Home
- >
- Networking
- >
- Legacy
- >
- Switches, Hubs, Modems
- >
- Re: HP2848 port asymmetry? MPI all-to-all communic...
Switches, Hubs, and Modems
1752796
Members
5938
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
тАО03-02-2006 07:44 PM
тАО03-02-2006 07:44 PM
HP2848 port asymmetry? MPI all-to-all communication
Hi,
I wonder whether the HP2848 switch has any asymmetry build-in (backplane?), or if any asymmetry could be induced accidentally (port buffers?) by the way in which it is used. I am doing MPI all-to-all communication on 32 ports and the performance differs depending on whether the MPI ranks are ordered with respect to the switch's ports or not.
In an ordered fashion, meaning that the process with MPI rank 0 uses port 0, rank 1 port 1 and so on, the performance is a lot worse for messages larger than approx. 2 k compared to the case when the ranks are shuffled (e.g. rank 0 on port 13, rank 1 on port 24, ...)
This behaviour is seen when using 32 ports, not for 16, 8, 4 ports. I tested with 3 different MPI implementations, so I think it is most likely due to the switch.
Performance differences are seen in the following way: For a given message size, a typical all-to-all call takes 0.03 seconds if the ranks are shuffled, but 0.2 seconds if the ranks are ordered. Doing the test several times with ranks that are slighly shuffled yields something like 0.03s, 0.20s, 0.21s, 0.03s, ... either fast or slow.
Does anyone have an idea what is the reason for this behaviour??
Thanks,
Carsten
I wonder whether the HP2848 switch has any asymmetry build-in (backplane?), or if any asymmetry could be induced accidentally (port buffers?) by the way in which it is used. I am doing MPI all-to-all communication on 32 ports and the performance differs depending on whether the MPI ranks are ordered with respect to the switch's ports or not.
In an ordered fashion, meaning that the process with MPI rank 0 uses port 0, rank 1 port 1 and so on, the performance is a lot worse for messages larger than approx. 2 k compared to the case when the ranks are shuffled (e.g. rank 0 on port 13, rank 1 on port 24, ...)
This behaviour is seen when using 32 ports, not for 16, 8, 4 ports. I tested with 3 different MPI implementations, so I think it is most likely due to the switch.
Performance differences are seen in the following way: For a given message size, a typical all-to-all call takes 0.03 seconds if the ranks are shuffled, but 0.2 seconds if the ranks are ordered. Doing the test several times with ranks that are slighly shuffled yields something like 0.03s, 0.20s, 0.21s, 0.03s, ... either fast or slow.
Does anyone have an idea what is the reason for this behaviour??
Thanks,
Carsten
3 REPLIES 3
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО03-04-2006 11:16 PM
тАО03-04-2006 11:16 PM
Re: HP2848 port asymmetry? MPI all-to-all communication
Hi Carsten,
The performance degradation is probably due to packet loss. This switch has a dynamic memory buffer pool shared across all ports. With 32 ports active, you probably are exceeding the buffers assigned to the default priority queues.
Try enabling 'qos-passthrough-mode'. This is designed to enhance the performance of line-rate traffic transfers. This feature is intended where QoS is not of major importance, but where lossless data transfers are key.
See 10-25 in the following guide:
ftp://ftp.hp.com/pub/networking/software/Mgmt-Oct2005-59906023-Chap10.pdf
Cheers
The performance degradation is probably due to packet loss. This switch has a dynamic memory buffer pool shared across all ports. With 32 ports active, you probably are exceeding the buffers assigned to the default priority queues.
Try enabling 'qos-passthrough-mode'. This is designed to enhance the performance of line-rate traffic transfers. This feature is intended where QoS is not of major importance, but where lossless data transfers are key.
See 10-25 in the following guide:
ftp://ftp.hp.com/pub/networking/software/Mgmt-Oct2005-59906023-Chap10.pdf
Cheers
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО03-08-2006 12:11 AM
тАО03-08-2006 12:11 AM
Re: HP2848 port asymmetry? MPI all-to-all communication
Hi Manfred,
thanks a lot for your reply. I enabled qos-passthrough-mode in the switch and it helped a bit, that means now the critical message size where packet drops occur is about twice as big. The fact that shuffling the input ports solved the problem caught my interest and I took a look into the switch.
It seems like four 12-port BCM5690 are connected to an 8-port 5670 with 10 Gbps link each. Probably here is the bottleneck: 12 ports generate more than 10 Gbps on these links during my test program, so the switch has to drop packets internally if they cannot be stored somewhere. When I only use 9 of each 12 ports (leaving the ports 10-12, 22-24,... empty) there is no packet loss. (Using 10 ports out of 12 still shows drops, don't know why).
So my conclusion is: When only 32 ports are needed, don't use ports 1-32 but rather 1-9, 13-21,25-33,37-45 to get maximum performance.
Or generally distribute the non-used ports evenly across the switch. Am I right here?
Thanks,
Carsten
thanks a lot for your reply. I enabled qos-passthrough-mode in the switch and it helped a bit, that means now the critical message size where packet drops occur is about twice as big. The fact that shuffling the input ports solved the problem caught my interest and I took a look into the switch.
It seems like four 12-port BCM5690 are connected to an 8-port 5670 with 10 Gbps link each. Probably here is the bottleneck: 12 ports generate more than 10 Gbps on these links during my test program, so the switch has to drop packets internally if they cannot be stored somewhere. When I only use 9 of each 12 ports (leaving the ports 10-12, 22-24,... empty) there is no packet loss. (Using 10 ports out of 12 still shows drops, don't know why).
So my conclusion is: When only 32 ports are needed, don't use ports 1-32 but rather 1-9, 13-21,25-33,37-45 to get maximum performance.
Or generally distribute the non-used ports evenly across the switch. Am I right here?
Thanks,
Carsten
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО03-08-2006 03:09 AM
тАО03-08-2006 03:09 AM
Re: HP2848 port asymmetry? MPI all-to-all communication
Hi Carsten,
Yes, you are correct. It definately sounds like the 12 ports are generating more than 10 Gbps for a sustained duration in your test configuration. Distributing the ports as you suggest will elimitate the 10 Gbps bottleneck to the 5670.
In "real-world" environments, most networks will not be subjected to this type of traffic. For high performance environments, you might want to consider our recently introduced 3500 switch that does not suffer from this limitation.
Cheers
Yes, you are correct. It definately sounds like the 12 ports are generating more than 10 Gbps for a sustained duration in your test configuration. Distributing the ports as you suggest will elimitate the 10 Gbps bottleneck to the 5670.
In "real-world" environments, most networks will not be subjected to this type of traffic. For high performance environments, you might want to consider our recently introduced 3500 switch that does not suffer from this limitation.
Cheers
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