jugglingdb

jugglingdb copied to clipboard

I'm trying to decide between sequelize and juggling db but I can't find to seem any performance data about jugglingdb. This makes deciding on whether to use this or not very difficult

Honestly, I don't think performance should be the key lens for looking at an ORM. Sequelize is certainly more popular, and seems to have gained many of the features that made me decide not to use it originally.

Shrugs

That said, if you do a comparison, please share.

performance issues with the code, while important, is meaningless, since it's entirely dependent on the database backend performance. testing the performance using a mocked database is just as useful and testing performance reassignment in for loops and finding out that one way works at 77.000.000 ops/sec and the other 72.000.000 ops/sec, while 5.000.000 seem a lot, hardly even you'll do this on a real world scenario.

A while ago I practiced benchmark driven development and optimized js based on memory usage / cpu usage graphs under benchmark function. All slow crap was removed and classes usage was rewritten. But I never compared with other ORMs because it doesn't make any sense for me - I only compare with pure driver to minimize overhead added by jugglingdb. And I agree - that comparison could be useful for others, so if you want to compare I would recommend you to use to write simple benchmark and draw graphs using R.

On Wed, Jun 26, 2013 at 5:54 AM, Paulo Cesar [email protected]:

performance issues with the code, while important, is meaningless, since it's entirely dependent on the database backend performance. testing the performance using a mocked database is just as useful and testing performance reassignment in for loops and finding out that one way works at 77.000.000 ops/sec and the other 72.000.000 ops/sec, while 5.000.000 seem a lot, hardly even you'll do this on a real world scenario.

— Reply to this email directly or view it on GitHubhttps://github.com/1602/jugglingdb/issues/298#issuecomment-20022780 .

I'm looking at creating a script for generating test results for jugglingdb, node-orm, and sequelize.. I think it might be useful in the future for any others who are simply curious about the numbers. Obviously I know that the numbers for orm versus no-orm don't tell you much but the performance differences between two ORMs might come in very handy

If you're going to do it, I'd curious about bookshelf (site, github), also.

Hey guys, I took this project; https://github.com/sequelize/sequelize-performance

Added support for juggling and brought all the other ORMs up to date.. How should I post this code? Or should I just put the results I have?

Maybe just submit pull request?

On Tue, Jul 2, 2013 at 6:30 PM, tolgaek [email protected] wrote:

Hey guys, I took this project; https://github.com/sequelize/sequelize-performance

Added support for juggling and brought all the other ORMs up to date.. How should I post this code? Or should I just put the results I have?

— Reply to this email directly or view it on GitHubhttps://github.com/1602/jugglingdb/issues/298#issuecomment-20348823 .

Thanks, Anatoliy Chakkaev

well, doing some tests, the major overhead of jugglingdb right now is the use of Function.bind as I created an issue in here #356. The problem lies with the V8 engine and not jugglingdb perse. when I wrote about the 5 million ops/sec difference, is actually a lot more when using Function.bind, and in a really busy server, it does make a difference.

I'm doing a similar thing here. These are just dumb insertion test, I'm not checking relations or other things. Just "overhead added on node-mysql" :)

If you have suggestion on things I can stress please let me know.