[Beowulf] hi, how to do clustering
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Mike Davis jmdavis at mail2.vcu.eduThu Aug 4 13:10:39 PDT 2005
- Previous message: [Beowulf] hi, how to do clustering
- Next message: [Beowulf] rsh message error
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Heartbeat solves half of the problem. The load balancing issue requires more work. There are a couple of ways to go with it. One is to use some additional hardware to loadbalance qmail across two nodes. The specific implementation of qmail will determine how complex this needs to be. Currently you are running one qmail server. Is the performance issue only during peak useage, or are you seeing performance issues throughout the day? Does the qmail server also serve as a mailserver for users or is it expressly smtp? Separating out the smtp from pop and imap can be a good strategy to improve performance. Mike Davis Lombard, David N wrote: >The clustering used by this group does not support the work you want to >do. You need to look for high-availability clustering; this list is >interested in high-performance computing, which a mail server is not... > >One suitable site is http://www.linux-ha.org/, in particular, the >heartbeat project. > > >
- Previous message: [Beowulf] hi, how to do clustering
- Next message: [Beowulf] rsh message error
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
