Re: Ignite Atomic Long

Hi, I'm using a wrapper to start the nodes as service. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Ignite-Atomic-Long-tp7706p7720.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Ignite Atomic Long

Hi, I'm having some trouble with atomic long inside my compute job. I'm getting the following exception once a while: FINEST|4532/0|Service AntheusMatchServerNode|16-09-13 10:51:23|[13:51:23,382][SEVERE][pub-#46%null%][GridCacheAtomicLongImpl] Failed to add and get: o.a.i.i.processors.datastru

Re: JVM tuning

Hi Alexey, thank you for your response. I've implemented your suggestions and now my cluster is with 100% cpu utilization. Best regards, Caio -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/JVM-tuning-tp7532p7705.html Sent from the Apache Ignite Users mailing

JVM tuning

As I've mentioned here , I'm trying to use Ignite within a biometric identification system to store fingerprint templates in memory and to use Ignite's compute grid capability to perform matching

Re: Access cache entries in C++ via JNI call

Hi Igor, my cluster has 34 server nodes and 1 client node. Biometric data (fingerprint templates) is spread across a distributed cache (about 50 million fingerprint templates). The client node creates an identification task, which has to compare an input fingerprint against all stored templates. T

Access cache entries in C++ via JNI call

Hi, I'm trying to implement a biometric identification system using Ignite. The first version was not good enough because in my job, for each cache entry, I was calling a C++ method through JNI to process that entry and the time to transfer all the data from Java to C++ was killing the performance