· Content
· News
· Articles
· Mailinglists
· Knowledgebase
· Trouble Tickets
· Files
· Glossary
· Links
· Compatibility Lists
· Forums
Welcome to our website
To take full advantage of all features you need to login or register. Registration is completely free and takes only a few seconds.
Improving Linux Scalability and Performance
Posted by: Hooz on: 01/19/2003 02:45 PM [ Print | 2 comment(s) ]
We all know that Linux has its benefits for some server roles, but it's been a difficult point to prove. The good folks over at IBM have taken it upon themselves to show just how good Linux is for things like web serving, and what it needs to improve on to step up to the enterprise level. There is a ton of good stuff in the article, so I'll just grab you a random quote:
Linux exhibits high reliability as a Web server operating system, and it has significant market share in this market. Web servers are typically low-end to midrange systems with up to 4-way symmetric multiple processors (SMP); enterprise-level systems have more complex requirements, such as larger processor counts and I/O configurations and significant memory and bandwidth requirements. In order for Linux to be enterprise-ready and commercially viable in the SMP market, its SMP scalability, disk and network I/O performance, scheduler, and virtual memory manager must be improved relative to commercial UNIX systems.
« Friday Night Retro Gaming! Quake2: Ooh Baby! · Improving Linux Scalability and Performance
· Intel VTune for Linux »
Comment
|
HFU SMP Guru Posts: 2000 Joined: 2002-06-15 |
There are many softwares now a day are optimized for Linux Platform. One of example would be Vmware, it runs guest OS faster on Linux Platform than Windows Platform under same hardware setup/configuration. I just wish Vmware could release a future version that will take advantage of SMP/SMT in the guest OS. HFU |
Comment
|
AntiBasic The Digital Janitor Posts: 271 Joined: 2002-09-12 |
This is pretty much just a rehashing of the 2001 FREENIX Scalable Linux Scheduling by Stephen Molloy and Peter Honeyman, CITI, University of Michigan at USENIX. You can find the abstract and full text here. The original paper is also available here but the title is slightly different: ELSC : Scalable Linux Scheduling on a Symmetric Multi-Processor Machine. Should better have been titled: Adapting BSD scheduling to Linux. :rolleyes: But it did point out some interesting ideas that Linux had in order to maintain processor affinity. Giving everyone an equal chance when they're clearly not equal is called what kids? COMMUNISM! |
































