Re: What is the most efficient way to scan all data partitions?

2018-05-09 Thread Dave Harvey
When running on AWS, I found that what the "disk" that you are writing to is
the most critical issue for Ignite.   EC2 instances with local SSDs have
about 20x the write rate as a multiple 3 TB GP2 problems,   and using actual
disks (e.g., EBS) for Ignite Persistence storage   is a non-starter.Once
you have the right storage, the other parameter settings can tweak your
performance higher.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/


Re: What is the most efficient way to scan all data partitions?

2018-05-09 Thread aealexsandrov
Hi,

It's normal that writing to disk is causing the degradation. You can do
several optimizations to speed up writting to disk:

https://apacheignite.readme.io/docs/durable-memory-tuning
https://apacheignite.readme.io/docs/performance-tips

However you can try to test your example with next DefaultDataRegion
configuration:



















You can also set walMode to NONE but in this case you can lose the data.

Also if you have more then one hard disk then you can think about separating
of the data and wal:

https://apacheignite.readme.io/docs/durable-memory-tuning#section-separate-disk-device-for-wal

According your example. It looks like ok but I don't see where you close
your streamer.

BR,
Andrei



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/