DLinq (Linq to SQL) Performance (Part 5)

re: DLinq (Linq to SQL) Performance Part 5

Roger Jennings — Mon, 16 Jul 2007 23:33:29 GMT

Rico,

Thanks for the update and conclusions. I'm eagerly awaiting the beta 2 bits to see if I can duplicate your 93% throughput.

--rj

http://oakleafblog.blogspot.com/2007/07/rico-mariani-dlinq-linq-to-sql.html

re: DLinq (Linq to SQL) Performance Part 5

ricom — Tue, 17 Jul 2007 01:19:53 GMT

I don't know whether or not this 93% number has any meaning. It's a measure of raw overhead in one test case (500 selects from the Northwinds Orders table) but it doesn't represent what people will actually see in real world cases.

It was great to drive performance work in DLinq because it removes non-DLinq effects but beyond that... does it mean anything?

You can get different numbers just by dialing the knobs as you see fit. More rows, less rows, more runs, less runs, add business logic, add database latency, etc.

re: DLinq (Linq to SQL) Performance Part 5

Joku — Tue, 17 Jul 2007 10:01:48 GMT

What's the perf like in comparison if you had what you in this case have in DB, as some just unserialized (list of objects or whatever)?

I could imagine a few cases where you might not need all the things full DB has to offer, but just the LINQ facilities over object and good perf.

Is there some other cost in place of step 3/4 when using LINQ with objects in memory?

LINQ to SQL and (micro) Performance

Mike Taulty's Blog — Tue, 17 Jul 2007 12:18:17 GMT

Also been catching up on Rico Mariani's notes on improvements to LINQ to SQL performance between the...

re: DLinq (Linq to SQL) Performance Part 5

ricom — Tue, 17 Jul 2007 18:13:26 GMT

This is just an analysis of Linq to SQL.

Linq to Objects has totally different characteristics, the way you access them looks the same but other than that the execution is basically entirely different. It sort of has to be.

re: DLinq (Linq to SQL) Performance Part 5

Henry Boehlert — Tue, 17 Jul 2007 22:45:08 GMT

Now, why don't we send the compiled query (or parts of it in the presence of joins) to the SQL Server and save the SQL parsing and parameter type conversion?

Izindaba #1

From the software development trenches — Fri, 20 Jul 2007 13:30:18 GMT

I try to publish a short, weekly roundup of news that focuses on .NET and MS development related content

re: DLinq (Linq to SQL) Performance (Part 5)

kevini — Thu, 02 Aug 2007 20:05:57 GMT

One question I have is with the push to using DLINQ, do you see any sort of tools to help aid in proactive query tuning?

I realize you can use SPs through DLINQ, but it seems some of the beauty is in the raw SQL htat can be generated through the expressions. But, it also seems that the only way to profile to check indices/perf in general is to run it and reactively verify the SQL2005 metrics for if an index should be present and using those post-execution methods.

Currently, a nice thing about using raw SPs for everything is that you can have that list of SPs and create testpasses, or automatically script out executions to verify that the script is 'sound'. And it sounds like a rush to use DLINQ for anything more than CRUD queries might remove that important level of pre-release analysis.

But maybe there is some way that when a project is built to use that Meta data that is generated in a similar fashion to build out 'test cases' or test scripts even to use in a proactive manner. Is that sort of work being addressed as far as you know?

MSDN events - more about LINQ, Silverlight and WCF

Contagious Curiosity — Mon, 13 Aug 2007 03:33:25 GMT

Did you attend one my events in August to September 2007 and want more information about the topics covered?

MSDN events - more about LINQ, Silverlight and WCF

Noticias externas — Mon, 13 Aug 2007 04:37:43 GMT

Did you attend one my events in August to September 2007 and want more information about the topics covered

Community Convergence XXIX

Charlie Calvert's Community Blog — Mon, 13 Aug 2007 09:36:20 GMT

There are several good new blogs from members of the community team. Nevertheless, the most important

Community Convergence XXIX

Charlie Calvert's Community Blog — Mon, 13 Aug 2007 09:39:48 GMT

There are several good new blogs from members of the Microsoft C# team. Nevertheless, the most important

Resources for MSDN Events - LINQ, WCF, Silverlight

Veni, Vidi, Velcro... — Mon, 13 Aug 2007 19:54:50 GMT

Technorati tags: Resources , MSDN events , Portland MSDN Event , WCF , LINQ , Silverlight Thanks for

LINQ to SQL in Beta 2

Daniel Moth — Wed, 15 Aug 2007 20:59:18 GMT

LINQ to SQL in Beta 2

Daniel Moth — Wed, 15 Aug 2007 20:59:48 GMT

LINQ to SQL in Beta 2

Daniel Moth — Thu, 16 Aug 2007 15:56:44 GMT

LINQ to SQL in Beta 2

LINQ to SQL - compiled queries (with working example for beta 2)

Ronan Geraghty's Weblog — Fri, 24 Aug 2007 22:28:00 GMT

I've been meaning to dig into LINQ performance for some time (actually since it came up during one of

LINQ to SQL - compiled queries (with working example for beta 2)

MSDN Ireland Blog — Fri, 24 Aug 2007 22:28:00 GMT

I've been meaning to dig into LINQ performance for some time (actually since it came up during one of

LINQ to SQL - compiled queries (with working example for beta 2)

MSDN Ireland Blog — Fri, 24 Aug 2007 22:29:45 GMT

I've been meaning to dig into LINQ performance for some time (actually since it came up during one of

LINQ to SQL - compiled queries (with working example for beta 2), performance

Ronan Geraghty's Weblog — Fri, 24 Aug 2007 22:49:52 GMT

I've been meaning to dig into LINQ performance for some time (actually since it came up during one of

Risorse su Linq to SQL

jankyBlog — Mon, 27 Aug 2007 10:58:29 GMT

Risorse su Linq to SQL

LINQ to SQL : Some of the best BLOGs

Wriju's BLOG — Thu, 01 Nov 2007 20:31:47 GMT

Some of the best blogs on LINQ to SQL I found are available for great learning, Scott Guthrie The Famous

DLinq (Linq to SQL) Performance (Part 1)

Rico Mariani's Performance Tidbits — Thu, 15 Nov 2007 04:21:21 GMT

[ By popular demand, here are links for all 5 parts in the series Part 1 , Part 2 , Part 3 , Part 4 ,