How many segments each shard has and what is the reason of running multiple
shards in one machine?
Alex.
-Original Message-
From: qibaoyuan
To: solr-user
Sent: Fri, Apr 19, 2013 12:26 am
Subject: Re: solr-cloud performance decrease day by day
there are 6 shards and they are
- which is why you want only a modest amount of heap
available above what the data needs after a few queries have loaded caches.
-- Jack Krupansky
-Original Message-
From: qibaoyuan
Sent: Friday, April 19, 2013 3:15 AM
To: solr-user@lucene.apache.org
Subject: solr-cloud performance
Thanks manu,i will check it.
在 2013-4-19，下午4:26，Manuel Le Normand 写道：
> Can happen for various reasons.
>
> Can you recreate the situation, meaning restarting the servlet or server
> would start with good qTime and decrease from that point? How fast does
> this happen?
>
> Start by monitoring t
Can happen for various reasons.
Can you recreate the situation, meaning restarting the servlet or server
would start with good qTime and decrease from that point? How fast does
this happen?
Start by monitoring the jvm process, with oracle visualVM for example.
Monitor for frequent garbage collect
there are 6 shards and they are in one machine,and the jvm param is very
big,the physical memory is 16GB,the total #docs is about 150k,the index size of
each shard is about 1GB.AND there is indexing while searching,I USE auto commit
each 10min.and the data comes about 100 per minutes.
在 2013
Could you give more info about your index size and technical details of
your machine? Maybe you are indexing more data day by day and your RAM
capability is not enough anymore?
2013/4/19 qibaoyuan
> Hello,
>i am using sold 4.1.0 and ihave used sold cloud in my product.I have
> found at first
Hello,
i am using sold 4.1.0 and ihave used sold cloud in my product.I have found
at first everything seems good,the search time is fast and delay is slow,but it
becomes very slow after days.does any one knows if there maybe some params or
optimization to use sold cloud?
