You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@ignite.apache.org by Zhengqingzheng <zh...@huawei.com> on 2016/08/04 08:08:36 UTC
question about ignite benchmark test
Hi there,
I am reading ignite benchmark test result from this address：http://www.gridgain.com/resources/benchmarks/ignite-vs-hazelcast-benchmarks
And I see there are some graph result labeled with descriptions like this: Graphs: 1 key<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/006-1.5.0.final-646926c3-tx-optimistic-put-1-backup/Results.html>, 2 keys<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/014-1.5.0.final-646926c3-tx-putAll-bs2-1-backup/Results.html>, 6 keys<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/015-1.5.0.final-646926c3-tx-putAll-bs-6-1-backup/Results.html>, 10 keys<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/016-1.5.0.final-646926c3-tx-putAll-bs-10-1-backup/Results.html>.
Can anyone gives me a rough description for the meaning of different keys in the test?
Best regards,
Kevin
Re: 答复: question about ignite benchmark test
Posted by vkulichenko <va...@gmail.com>.
Kevin, Jason,
Here is the source code for these benchmarks:
https://github.com/apacheignite/yardstick-ignite. Most of the mentioned
parameters are configurable, but the published benchmarks are executed with
defaults unless otherwise specified.
-Val
--
View this message in context: http://apache-ignite-users.70518.x6.nabble.com/question-about-ignite-benchmark-test-tp6737p6812.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.
答复: question about ignite benchmark test
Posted by Zhengqingzheng <zh...@huawei.com>.
Hi Val,
Thank you for your reply.
One more question.
How about keys in SQL benchmark? Is it the parameter number or the return result of sql query?
Best regards,
Kevin
发件人: Vladimir Ozerov [mailto:vozerov@gridgain.com]
发送时间: 2016年8月4日 17:39
收件人: user@ignite.apache.org
主题: Re: question about ignite benchmark test
Hi Kevin,
This is the number of keys involved into PUT or GET operation. If there is 1 key, then IgniteCache.get() or IgniteCache.put() operation was used. If there are more keys, then we benchmarked IgniteCache.getAll() or IgniteCache.putAll() operations.
On Thu, Aug 4, 2016 at 11:08 AM, Zhengqingzheng <zh...@huawei.com>> wrote:
Hi there,
I am reading ignite benchmark test result from this address：http://www.gridgain.com/resources/benchmarks/ignite-vs-hazelcast-benchmarks
And I see there are some graph result labeled with descriptions like this: Graphs: 1 key<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/006-1.5.0.final-646926c3-tx-optimistic-put-1-backup/Results.html>, 2 keys<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/014-1.5.0.final-646926c3-tx-putAll-bs2-1-backup/Results.html>, 6 keys<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/015-1.5.0.final-646926c3-tx-putAll-bs-6-1-backup/Results.html>, 10 keys<http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/016-1.5.0.final-646926c3-tx-putAll-bs-10-1-backup/Results.html>.
Can anyone gives me a rough description for the meaning of different keys in the test?
Best regards,
Kevin
Re: question about ignite benchmark test
Posted by Jason <fq...@outlook.com>.
hi Vladimir,
What's the size of the key and value in the testing?
For the Atomic 10keys, the total entries during the 5 mins' testing is ~35K
* 10 * 300 ~= 100M. What's the eviction policy and how many entries are kept
in the memory during the whole testing?
BTW, is there any benchmark for the big cache in off heap, say over 20G per
node and hundreds of gigabytes or over 1T in total ?
Thanks,
-Jason
--
View this message in context: http://apache-ignite-users.70518.x6.nabble.com/question-about-ignite-benchmark-test-tp6737p6753.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.
Re: question about ignite benchmark test
Posted by Vladimir Ozerov <vo...@gridgain.com>.
Hi Kevin,
This is the number of keys involved into PUT or GET operation. If there is
1 key, then *IgniteCache.get() *or *IgniteCache.put() *operation was used.
If there are more keys, then we benchmarked *IgniteCache.getAll() *or
*IgniteCache.putAll()
*operations.
On Thu, Aug 4, 2016 at 11:08 AM, Zhengqingzheng <zh...@huawei.com>
wrote:
> Hi there,
>
> I am reading ignite benchmark test result from this address：
> http://www.gridgain.com/resources/benchmarks/ignite-vs-hazelcast-benchmarks
>
> And I see there are some graph result labeled with descriptions like this:
> Graphs: 1 key
> <http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/006-1.5.0.final-646926c3-tx-optimistic-put-1-backup/Results.html>
> , 2 keys
> <http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/014-1.5.0.final-646926c3-tx-putAll-bs2-1-backup/Results.html>
> , 6 keys
> <http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/015-1.5.0.final-646926c3-tx-putAll-bs-6-1-backup/Results.html>
> , 10 keys
> <http://www.gridgain.com/wp-content/benchmarks/ignite-vs-hazelcast/tx-put/016-1.5.0.final-646926c3-tx-putAll-bs-10-1-backup/Results.html>
> .
>
> Can anyone gives me a rough description for the meaning of different keys
> in the test?
>
>
>
>
>
> Best regards,
>
> Kevin
>
>
>
>
>