Open source version slow on purpose?
Closed this issue · 7 comments
Hi
Your chart is interesting, normally enterprise means extra features, etc, however performance should be the same, how is it that its 10 times slower.
Hazelcast and others offer the same framework and charge for support and enterprise specific features... same with gridgain
thanks
Paying clients often have specific actions they need to be fast. We can
optimise and test those actions to ensure they deliver the performance they
need.
However we don't have performance tests for every possible combination of
setting and action so if you want support for free we don't look to achieve
as agreesive performance targets, we only try to be faster than our
competition.
The numbers are the treshold at which we consider it a bug. The actual
performance is likely to be different and these numbers most likely need
revision as we have made significant improvements since then.
Regards, Peter.
On 17 Jul 2016 21:48, "Serge Bornow" notifications@github.com wrote:
Hi
Your chart is interesting, normally enterprise means extra features, etc,
however performance should be the same, how is it that its 10 times slower.
Hazelcast and others offer the same framework and charge for support and
enterprise specific features... same with gridgain
thanks—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#9, or mute the thread
https://github.com/notifications/unsubscribe-auth/ABBU8Wc5InsULChelYgvu7a2wWTJ3hwjks5qWpUMgaJpZM4JOTzb
.
We don't slow down the open source very but we don't claim that for
anything you need it to do we will fix the performance for you as much as
if you pay for support.
On 17 Jul 2016 22:08, "Peter Lawrey" peter.lawrey@chronicle.software
wrote:
Paying clients often have specific actions they need to be fast. We can
optimise and test those actions to ensure they deliver the performance they
need.
However we don't have performance tests for every possible combination of
setting and action so if you want support for free we don't look to achieve
as agreesive performance targets, we only try to be faster than our
competition.
The numbers are the treshold at which we consider it a bug. The actual
performance is likely to be different and these numbers most likely need
revision as we have made significant improvements since then.Regards, Peter.
On 17 Jul 2016 21:48, "Serge Bornow" notifications@github.com wrote:
Hi
Your chart is interesting, normally enterprise means extra features, etc,
however performance should be the same, how is it that its 10 times slower.
Hazelcast and others offer the same framework and charge for support and
enterprise specific features... same with gridgain
thanks—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#9, or mute the
thread
https://github.com/notifications/unsubscribe-auth/ABBU8Wc5InsULChelYgvu7a2wWTJ3hwjks5qWpUMgaJpZM4JOTzb
.
Thank you for quick response Peter
However it does state in readme:
Chronicle Enterprise will be tested to a higher spec and will have more heavily optimized components.
So that still means oss version is slower on purpose.
From what i understand, if someone contributes to Gridgain or hazelcast or others, that benefits them as much as they contribute to oss or community version and making one substantially slower isn't a good OSS practice (unless you're oracle).
cheers
I guess I don't understand how you come to that conclusion. Like I said
the code is the same.
The performance is the same for an given operation, the only difference is
what a developer can expect us to fix depending on whether they pay for it
or not. However, once fixed, that improvement is available to everyone.
On 18 July 2016 at 14:44, Serge Bornow notifications@github.com wrote:
Thank you for quick response Peter
However it does state in readme:
Chronicle Enterprise will be tested to a higher spec and will have more heavily
optimized components.So that still means oss version is slower on purpose.
From what i understand, if someone contributes to Gridgain or hazelcast or
others, that benefits them as much as they contribute to oss or community
version and making one substantially slower isn't a good OSS practice
(unless you're oracle).cheers
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#9 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABBU8bIzsLH7CkVmVbCJODZWh54TGVWfks5qW4M1gaJpZM4JOTzb
.
We don't have an OSS version and a commercial version of the code.
On 18 July 2016 at 15:37, Peter Lawrey peter.lawrey@chronicle.software
wrote:
I guess I don't understand how you come to that conclusion. Like I said
the code is the same.
The performance is the same for an given operation, the only difference is
what a developer can expect us to fix depending on whether they pay for it
or not. However, once fixed, that improvement is available to everyone.On 18 July 2016 at 14:44, Serge Bornow notifications@github.com wrote:
Thank you for quick response Peter
However it does state in readme:
Chronicle Enterprise will be tested to a higher spec and will have more heavily
optimized components.So that still means oss version is slower on purpose.
From what i understand, if someone contributes to Gridgain or hazelcast
or others, that benefits them as much as they contribute to oss or
community version and making one substantially slower isn't a good OSS
practice (unless you're oracle).cheers
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#9 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABBU8bIzsLH7CkVmVbCJODZWh54TGVWfks5qW4M1gaJpZM4JOTzb
.
Ahhh, ok i was a bit confused by that point on readme.
Thanks for confirming that it is the same. It just seemed like oss isnt optimized but enterprise is.
That is good feedback. I will update the documentation to make this clearer.
On 18 July 2016 at 16:12, Serge Bornow notifications@github.com wrote:
Ahhh, ok i was a bit confused by that point on readme.
Thanks for confirming that it is the same. It just seemed like oss isnt
optimized but enterprise is.—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#9 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABBU8Wlob7amSUFK7Nd7WgWkB5qtwmbdks5qW5fUgaJpZM4JOTzb
.