HPE GreenLake Administration
- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- L1000 system performance mystery
Operating System - HP-UX
1830508
Members
2477
Online
110006
Solutions
Forums
Categories
Company
Local Language
back
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Forums
Discussions
Discussions
Discussions
Forums
Discussions
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
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
12-01-2004 09:44 AM
12-01-2004 09:44 AM
L1000 system performance mystery
We have an L1000 with 1 CPU and 1 GB of RAM running Progress. The following was found while analyzing data gathered with "vmstat 5" over the course of 1.5 hours:
-There is no memory bottleneck.
-There is no I/O bottleneck.
The following started at 10:33 AM and continued until the data collection stopped:
-The load average dropped from 4.6 to 4.3
-CPU idle time dropped from 60-100% down to a flat 0 and stayed there
-The CPU was in user mode for 80-90% of the time (up from 0-30%)
The CPU was in system mode for 10-20% of the time (up from 0-5%)
-Paging activity was light with 0 pageouts
-No more than 1 blocked process at any one time (same as before 10:33)
Any idea what can cause the load average to remain relatively unchanged yet the CPU idle time to drop to a flat zero?
-There is no memory bottleneck.
-There is no I/O bottleneck.
The following started at 10:33 AM and continued until the data collection stopped:
-The load average dropped from 4.6 to 4.3
-CPU idle time dropped from 60-100% down to a flat 0 and stayed there
-The CPU was in user mode for 80-90% of the time (up from 0-30%)
The CPU was in system mode for 10-20% of the time (up from 0-5%)
-Paging activity was light with 0 pageouts
-No more than 1 blocked process at any one time (same as before 10:33)
Any idea what can cause the load average to remain relatively unchanged yet the CPU idle time to drop to a flat zero?
1 REPLY 1
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
12-01-2004 10:12 AM
12-01-2004 10:12 AM
Re: L1000 system performance mystery
Vmstat is not my weapon of choice for this; about the only column of vmstat that I ever pay attention to is the page-out rate. Glance is a much, much better tool for this and you could drill down into the busy process. You can install the fully-functional Trial Version from any application CD set.
The high fraction of user mode really suggests that this is probably application code that is CPU intensive. Possibly very tight loops or stored procedure executions -- but these are just guesses. What has changed recently in the application code and/or SQL?
One other possibility (since this is database stuff) is very high logical i/o (either in UNIX buffer cache or in the database cache) rates but with low physical i/o rates -- because most (or all) of the data are cached. Glance can pinpoint this sort of thing very quickly.
The high fraction of user mode really suggests that this is probably application code that is CPU intensive. Possibly very tight loops or stored procedure executions -- but these are just guesses. What has changed recently in the application code and/or SQL?
One other possibility (since this is database stuff) is very high logical i/o (either in UNIX buffer cache or in the database cache) rates but with low physical i/o rates -- because most (or all) of the data are cached. Glance can pinpoint this sort of thing very quickly.
If it ain't broke, I can fix that.
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.
Company
Events and news
Customer resources
© Copyright 2025 Hewlett Packard Enterprise Development LP