Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Plain View
Accumulo >> mail # user >> Which key/value pairs are held in memory?


+
Kristopher Kane 2012-11-01, 11:34
+
Josh Elser 2012-11-01, 11:55
Copy link to this message
-
Re: Which key/value pairs are held in memory?
On Thu, Nov 1, 2012 at 4:34 AM, Kristopher Kane <[EMAIL PROTECTED]>wrote:

> In the manual, it is stated that a write is sent to multiple loggers,
> written to the wa log and key/value pairs stored in a memtable and an
> index.
>
> Since the write hits several servers for resiliency, does each of the
> servers that receives the write place the same data in memory or do
> they each place a different portion in memory such that a read could
> span multiple servers for a performance gain?
>

Reads only use a single server, so only the tablet server hosting the
tablet puts the data in memory.  On other machines it is just logged to
disk.

Billie

>
> Thanks,
>
> -Kris
>
+
Kristopher Kane 2012-11-01, 13:57
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB