You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tinkerpop.apache.org by Jeffrey Freeman <je...@syncleus.com> on 2015/07/13 19:36:09 UTC

An update on Tinkerpop OGM/ORM Benchmarks including Tinkerpop3

Hey guys. You might recall my post from a little while back comparing the
various ORM/OGM including benchmarks of them all (Totorom, Frames, Ferma,
Blueprints, and Gremlin). Well thanks to the contribution of the Peapod
developer we now also expanded these benchmarks to compare against Peapod
and Tinkerpop3, So I wanted to share the results with everyone here.

The results for Tinkerpop3 were less than promising showing a huge
performance hit in the new version over the old something on the order of
10x slower or more. Since Peapod is built on Tinkerpop3 it suffers the same
poor performance (and adds some overhead of its own of course).

The full results can be viewed on this page, it also includes a link to the
GitHub project hosting the code for the benchmarks:
http://wiki.syncleus.com/index.php/Ferma:Comparing_the_Alternatives

Re: An update on Tinkerpop OGM/ORM Benchmarks including Tinkerpop3

Posted by Jeffrey Phillips Freeman <fr...@gmail.com>.
In case it is of any use to you guys (I just saw these messages now, I know 
its a year late)... 1) Ferma is now on an Apache v2 license. So no worries 
there. 2) I havent done recent TP3 benchmarks, anyone have any they can 
point me to so I can evaluate if i can start recommending TP3 yet?

On Monday, 27 July 2015 09:10:23 UTC-4, Rômulo Victor wrote:
>
> Hi Jeffrey,
> I have a question concerning the license used by the Ferma project, which 
> is OSCL type C. From what I've read here 
> <http://wiki.syncleus.com/index.php/Syncleus:Open_Source_Community_License_-_Type_C>, 
> it is more restrictive than Apache License but less restrictive than GPL. 
> Is that correct? Could you please explain how they differ?
> Thanks for your attention.
>
> Em segunda-feira, 13 de julho de 2015 14:36:23 UTC-3, Jeffrey Freeman 
> escreveu:
>>
>> Hey guys. You might recall my post from a little while back comparing the 
>> various ORM/OGM including benchmarks of them all (Totorom, Frames, Ferma, 
>> Blueprints, and Gremlin). Well thanks to the contribution of the Peapod 
>> developer we now also expanded these benchmarks to compare against Peapod 
>> and Tinkerpop3, So I wanted to share the results with everyone here.
>>
>> The results for Tinkerpop3 were less than promising showing a huge 
>> performance hit in the new version over the old something on the order of 
>> 10x slower or more. Since Peapod is built on Tinkerpop3 it suffers the same 
>> poor performance (and adds some overhead of its own of course).
>>
>> The full results can be viewed on this page, it also includes a link to 
>> the GitHub project hosting the code for the benchmarks:
>> http://wiki.syncleus.com/index.php/Ferma:Comparing_the_Alternatives
>>
>

Re: An update on Tinkerpop OGM/ORM Benchmarks including Tinkerpop3

Posted by Rômulo Victor <rb...@gmail.com>.
Hi Jeffrey,
I have a question concerning the license used by the Ferma project, which 
is OSCL type C. From what I've read here 
<http://wiki.syncleus.com/index.php/Syncleus:Open_Source_Community_License_-_Type_C>, 
it is more restrictive than Apache License but less restrictive than GPL. 
Is that correct? Could you please explain how they differ?
Thanks for your attention.

Em segunda-feira, 13 de julho de 2015 14:36:23 UTC-3, Jeffrey Freeman 
escreveu:
>
> Hey guys. You might recall my post from a little while back comparing the 
> various ORM/OGM including benchmarks of them all (Totorom, Frames, Ferma, 
> Blueprints, and Gremlin). Well thanks to the contribution of the Peapod 
> developer we now also expanded these benchmarks to compare against Peapod 
> and Tinkerpop3, So I wanted to share the results with everyone here.
>
> The results for Tinkerpop3 were less than promising showing a huge 
> performance hit in the new version over the old something on the order of 
> 10x slower or more. Since Peapod is built on Tinkerpop3 it suffers the same 
> poor performance (and adds some overhead of its own of course).
>
> The full results can be viewed on this page, it also includes a link to 
> the GitHub project hosting the code for the benchmarks:
> http://wiki.syncleus.com/index.php/Ferma:Comparing_the_Alternatives
>