Hello! I think you should expect that. When you do more parallel operations, there's more locking and throughput decreases.
However, we don't know enough about your specific workload to really understand what's going on. Regards, -- Ilya Kasnacheev вт, 27 нояб. 2018 г. в 16:24, Naveen <[email protected]>: > These are settings we have > > maxSize=#{80L * 1024 * 1024 * 1024} > checkpointPageBufferSize=#{8L * 1024 * 1024 * 1024} > pageSize=#{4 * 1024} > > <property name="writeSynchronizationMode" > value="PRIMARY_SYNC " /> > > > Any clues we have for this dip in TPS > > Thanks > Naveen > > > > -- > Sent from: http://apache-ignite-users.70518.x6.nabble.com/ >
