- Community Home
- >
- Servers and Operating Systems
- >
- Legacy
- >
- Operating System - Tru64 Unix
- >
- Kernel tuning - round_robin & Shed_min_idle
Operating System - Tru64 Unix
1752795
Members
6249
Online
108789
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
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
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
05-04-2003 02:49 AM
05-04-2003 02:49 AM
Kernel tuning - round_robin & Shed_min_idle
I've got ES45, 4x1GHz, 16MB RAM, Tru64 5.1A PK3, Oracle 8.1.7.4, Sql client app called Concorde(By Damgaard -> Navision -> MicroSoft).
System load is CPU-intensive - IO bandwith looks enough.
Lots of small&simple SQL between Oracle and Concorde.
In the:
> man sys_attrs_proc
I've found 2 interesting parameters -
round_robin_switch_rate and sched_min_idle
[man quoting is skipped...]
1) What is CPU overhead on context switching?
2) When task (Oracle's dedicated process or SQL client) done it's job and sent query/reply to it's peer - the rest of timeslice is used or wasted? (Or, can I allocate BIG timeslice).
3) Can I obtain CPU cache hit rate?
4) Did someone played with this parameters, if there is any effect?
And, more:
5) Memory can be non- 2-way, 4-way interleaved.
Did anyone noticed performance difference?
6) Does Oracle 8.1.7 work on Marvel - ES80, ES1280? Anyone tryed?
7) Is XMesh utility included into Tru64 5.1B?
8) Does XMesh utility operate on Alpha ev6 systems?
System load is CPU-intensive - IO bandwith looks enough.
Lots of small&simple SQL between Oracle and Concorde.
In the:
> man sys_attrs_proc
I've found 2 interesting parameters -
round_robin_switch_rate and sched_min_idle
[man quoting is skipped...]
1) What is CPU overhead on context switching?
2) When task (Oracle's dedicated process or SQL client) done it's job and sent query/reply to it's peer - the rest of timeslice is used or wasted? (Or, can I allocate BIG timeslice).
3) Can I obtain CPU cache hit rate?
4) Did someone played with this parameters, if there is any effect?
And, more:
5) Memory can be non- 2-way, 4-way interleaved.
Did anyone noticed performance difference?
6) Does Oracle 8.1.7 work on Marvel - ES80, ES1280? Anyone tryed?
7) Is XMesh utility included into Tru64 5.1B?
8) Does XMesh utility operate on Alpha ev6 systems?
The fire follows shedule...
1 REPLY 1
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-06-2003 12:01 AM
05-06-2003 12:01 AM
Re: Kernel tuning - round_robin & Shed_min_idle
1) What is CPU overhead on context switching?
A context switch is not overhead its is necessary because either the current thread is done with the CPU for now or another thread needs the CPU more. If a thread finishes its time quantum and there is nothing else waiting to run then the scheduler will let the thread run for another quantum, no contect switch will take place.
2) When task (Oracle's dedicated process or SQL client) done it's job and sent query/reply to it's peer - the rest of timeslice is used or wasted?
The unused quantum is given to another thread.
3) Can I obtain CPU cache hit rate?
No.
4) Did someone played with this parameters, if there is any effect?
The amount of benefit depends on the app and the loading of the system. Only a real life test can determine if it will work for you.
5) Memory can be non- 2-way, 4-way interleaved.
Did anyone noticed performance difference?
Depends on the app. It can be very important if the app is memory intensive.
6) Does Oracle 8.1.7 work on Marvel - ES80, ES1280? Anyone tryed?
Yes.
7) Is XMesh utility included into Tru64 5.1B?
Yes.
8) Does XMesh utility operate on Alpha ev6 systems?
No.
A context switch is not overhead its is necessary because either the current thread is done with the CPU for now or another thread needs the CPU more. If a thread finishes its time quantum and there is nothing else waiting to run then the scheduler will let the thread run for another quantum, no contect switch will take place.
2) When task (Oracle's dedicated process or SQL client) done it's job and sent query/reply to it's peer - the rest of timeslice is used or wasted?
The unused quantum is given to another thread.
3) Can I obtain CPU cache hit rate?
No.
4) Did someone played with this parameters, if there is any effect?
The amount of benefit depends on the app and the loading of the system. Only a real life test can determine if it will work for you.
5) Memory can be non- 2-way, 4-way interleaved.
Did anyone noticed performance difference?
Depends on the app. It can be very important if the app is memory intensive.
6) Does Oracle 8.1.7 work on Marvel - ES80, ES1280? Anyone tryed?
Yes.
7) Is XMesh utility included into Tru64 5.1B?
Yes.
8) Does XMesh utility operate on Alpha ev6 systems?
No.
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