On 12/27/06, Biggy <[EMAIL PROTECTED]> wrote:
Does anyone know how/if Solr can handle say 30 server with 3 requests/sec ?
We've only gone to 10 replicated search servers at CNET (and that number turned out to be way overkill). For most types of requests, a single server can handle way more than 3 requests per second. Much depends on the types of queries you will be throwing at it, and the size and nature of your corpus.
The replication strategy i have uses JGroups which is unfortunately is unreliable. How does Solr replicate data across servers for instance ?
Searchers use rsync to pull index changes from the master. It takes a little work to set up (ssh config, etc), but the upside is that it's all fairly transparent. -Yonik
