ignite-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tolga Kavukcu <kavukcu.to...@gmail.com>
Subject Re: Blocked get operation problem
Date Thu, 24 Nov 2016 07:10:35 GMT
Hi Val,

Yes i access same key from all threads, even from all nodes probably 60000
access per second. For now i can't reach this tps with 56core x 3 server =
168 core. All get operations are made from servers. Seems like locking the
keys causes the problem. Cus i see lots of threads waiting for a lock on
get operation.

There are ~100 entries in the cache. This is our configuration. Can you see
any improvements on configuration.

<bean class="org.apache.ignite.configuration.CacheConfiguration" name="DEFAULT">
    <property name="atomicityMode" value="ATOMIC" />
    <property name="cacheMode" value="LOCAL" />
    <property name="memoryMode" value="OFFHEAP_TIERED" />
    <property name="affinity">
        <bean class="org.apache.ignite.cache.affinity.fair.FairAffinityFunction">
            <constructor-arg index="0" type="int" value="192"/>
    <!-- max memory is set to 0 means unlimited -->
    <property name="offHeapMaxMemory" value="0" />
    <property name="writeThrough" value="false" />
    <property name="writeBehindEnabled" value="false" />
    <property name="eagerTtl" value="false"/>

Previous provider was Chronicle-Map.


On Wed, Nov 23, 2016 at 9:24 PM, vkulichenko <valentin.kulichenko@gmail.com>

> Tolga,
> Are you accessing the same key all the time? How many entries do you have
> in
> cache?
> What cache provider did you use before?
> -Val
> --
> View this message in context: http://apache-ignite-users.
> 70518.x6.nabble.com/Blocked-get-operation-problem-tp9133p9158.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.



View raw message