- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - Linux
- >
- Re: vmstat
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
тАО01-27-2009 02:58 AM
тАО01-27-2009 02:58 AM
please correct me if I am wrong
in vmstat output, under the CPU column
'us' and 'sy'
man vmstat
us: Time spent running non-kernel code. (user time, including nice time)
sy: Time spent running kernel code. (system time)
i did a small test
# for i in `seq 1 300`; do oowriter; done
the value of 'us' increases
but when i did
# dd if=/dev/urandom of=dd.img bs=1K count=10000
the value of 'sy' increases
Question: whats the non-kernel code ?
My view/knowledge:
software/applications(e.g OpenOffice, gimp, email clients etc) and daemons(web/dhcp/dns/mail), databases(oracle.mysql etc) are non-kernel code ?
Question: what are the running kernel code ?
My view/knowledge:
if I save some text using OpenOffice, then OpenOffice instructs the kernel(Linux) to do the work then the 'sy' value will increase ? am i right ?
i.e if execute some apps/db/daemons then the time to load/execute thos software/apps/daemons will be reflected by the 'us'
and
if the software do some processing(calculation, and manipulation, creation and deletion of files, and saving in the file) then the vlaue will be reflected in 'sy' column ?
regards
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО01-27-2009 07:21 AM
тАО01-27-2009 07:21 AM
Re: vmstat
Yes.
>>> if I save some text using OpenOffice, then OpenOffice instructs the kernel(Linux) to do the work then the 'sy' value will increase ? am i right ?
Yes
The sy will be increased when the program ask to the kernel to do part of the job, normally, through a system call:
Swapping is a process that will use high sy time, because it has to deal with pages allocation/deallocation and device handling.
http://docs.cs.up.ac.za/programming/asm/derick_tut/syscalls.html
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО01-27-2009 09:12 PM
тАО01-27-2009 09:12 PM
Re: vmstat
Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО01-30-2009 03:17 AM
тАО01-30-2009 03:17 AM
Re: vmstat
how do I know the reason behind the high value of wa, i.e, is it because the high network I/O
or high disk I/O
I know the high value of 'bi' and 'bo'(disk I/O) is also the cause of high value of 'wa', but how do I know that if the high rate of network I/O is also involved in the high vlaues of 'wa'
Regard
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО01-30-2009 04:36 AM
тАО01-30-2009 04:36 AM
SolutionYou should monitor your network traffic with iptraf, or maybe, install collectl (collect for linux). It's an excelent performance tool.