Here is the PR https://github.com/apache/ignite/pull/409 - not ready to
merge, but cache full api suite passes, so I can run distributed benchmarks
tomorrow.
Thanks!
--
Yakov Zhdanov, Director R&D
*GridGain Systems*
www.gridgain.com
2016-01-16 19:04 GMT+03:00 Dmitriy Setrakyan :
> Yakov,
>
> Any
Yakov,
Any chance you can rerun the benchmarks to confirm your findings?
Would also be nice if someone else could run them as well. Is there a
branch that we can check out?
D.
On Fri, Jan 15, 2016 at 12:17 PM, Vladimir Ozerov
wrote:
> I also hardly believe that such change can give us such bi
I also hardly believe that such change can give us such big immediate
performance benefit. Profiling shows that in our standard benchmarks
futures doesn't generate so much garbage to get ~5% from field removal. I
would rather re-check if benchmark is valid.
On the other hand, lots of our utility c
If it really can give as 5% of performance we have to do it.
If anyone uses this methods we can support it by user request (if user
explicitly asked about) and it will work slower in that case. For example,
we can add *IgniteCache.withAsync(boolean useFutureWithTimer)* that will
give to user an as
All of optimizations applied to futures so far were extremely effective.
Ignite can create different number of futures per operation depending on
context. Multiple this by ops/sec.. This is probably one of the most
intensively instantiated (and therefore GCed) object. Internal futures is
very sensi
Really hard to believe that removing 16 bytes in futures gives 5% of
performance. Yakov, are you certain about this?
If this turns out to be true, let’s see if we can slim down the futures
used internally, without breaking the public API.
D.
On Thu, Jan 14, 2016 at 8:36 AM, Vladimir Ozerov
wrot
+1
BTW, corresponding ticket already exists. You can find it under umbrella
ticket IGNITE-2232
14 янв. 2016 г. 18:40 пользователь "Yakov Zhdanov"
написал:
> Guys,
>
> We have startTime(), duration() and endTime() methods which have several
> usages each along internals of the project, but to sup
Guys,
We have startTime(), duration() and endTime() methods which have several
usages each along internals of the project, but to support these methods we
have 2 long fields in GridFutureAdapter which gives us 16 bytes.
Other fields - res (reference 8 bytes at max), ignoreInterrupts (boolean 1
by
