This thread looks to be a little on the old side and therefore may no longer be relevant. Please see if there is a newer thread on the subject and ensure you're using the most recent build of any software if your question regards a particular product.
This thread has been locked and is no longer accepting new posts, if you have a question regarding this topic please email us at support@mindscape.co.nz
|
Hey, I am new to LightSpeed and have done an exhaustive search of the web and cannot make up my mind. What is faster: (SQL with LINQ) or (MySQL with LightSpeed)? No bench marks seem to perform cross platform benchmarking. Do you have any internal benchmarking tests you can share? DBLinq Vs Lightspeed Vs NHibernate? Thanks, Jared
|
|
|
|
|
Hey, I found an independent one, it uses LightSpeed 2.2 for the benchmarks though - do you expect speed improvements in version 3.0? LightSpeed 2.2 stacks up well with "Materialisation" and "Fetching" but below average in Queries - was the test done correctly? Can it be sped up in line with other ORMs? Does v3 support compiled queries? Thanks Jared |
|
|
|
|
Hello Jared, No, sorry, we don't have any benchmarks comparing LINQ to SQL with LightSpeed over MySQL, nor comparing LightSpeed with NHibernate. Regarding MySQL, this may depend on the queries: we've been told that for some queries the SQL we generate hits a performance issue in MySQL (basically we generate subselects rather than joins, and MySQL doesn't handle subselects as efficiently as it handles joins). This is on our list of things to look at but we can't commit to it right now. Regarding NHibernate, a customer did some tests comparing LightSpeed's LINQ provider with NHibernate's and posted the overall result at http://www.mindscape.co.nz/forums/Thread.aspx?ThreadID=2464, but we don't have broken-out details of the tests or results. Regarding 2.2 vs. 3.0, our smoke testing shows a small improvement in querying performance and equivalent saving (insert, update, delete) performance. These tests were conducted with SQL Server, but I would expect the same kind of profile with MySQL. The ORM Battle benchmarks are considered extremely controversial in the ORM community, partly because of the way they were done and partly because the operators of the ORM Battle site are not independent -- they make their own ORM. See the comment thread to http://ayende.com/blog/archive/2009/08/15/benchmarks-are-useless-yes-again.aspx for a lengthy (and, I'm afraid, ill-tempered) discussion between the site operator and a couple of ORM developers. Specifically regarding the LightSpeed query performance reported on that site, we do not believe the test was done correctly as we see much better performance than they do, but I don't think we've analysed their benchmarks well enough to understand what problem they ran into. V3 doesn't support compiled queries, though this is a frequently requested feature and is on the wishlist for post-3.0 -- but again I can't promise anything. |
|
|
|
|
Thank you for the reply Ivan, Yes I later realised they weren't independent! Oh well. After about a week back and forth I decided to go with LightSpeed anyway - it looks the best combination of Speed-Ease-Usability(Partial Classes)-Advancements(.NET4)-DB Compatibility. I think you guys are a safe choice going forward =) Thanks Jared |
|
|
|
