Challenge: Premature optimization, and all of that…
Let us look at the following pieces of code:
public void Consume(MyBooksRequest message) { var user = session.Get<User>(message.UserId); bus.Reply(new MyBooksResponse { UserId = message.UserId, Timestamp = DateTime.Now, Books = user.CurrentlyReading.ToBookDtoArray() }); } public void Consume(MyQueueRequest message) { var user = session.Get<User>(message.UserId); bus.Reply(new MyQueueResponse { UserId = message.UserId, Timestamp = DateTime.Now, Queue = user.Queue.ToBookDtoArray() }); } public void Consume(MyRecommendationsRequest message) { var user = session.Get<User>(message.UserId); bus.Reply(new MyRecommendationsResponse { UserId = message.UserId, Timestamp = DateTime.Now, Recommendations = user.Recommendations.ToBookDtoArray() }); }
Looking at this, I see that I have a requirement to getting my books, my queues and my recommendations. It appears that getting each datum is going to result in 2 queries, the first to load the User, and the second to lazy load the actual collection that we want to return.
An almost trivial optimization would be to eliminate the lazy loading, right? That would reduce the cost from 6 queries to just 3.
However, that assumption would be wrong. The following client code:
bus.Send(
new MyBooksRequest
{
UserId = userId
},
new MyQueueRequest
{
UserId = userId
},
new MyRecommendationsRequest
{
UserId = userId
});
Produces this SQL:
-- statement #1 enlisted session in distributed transaction with isolation level: Serializable -- statement #2 SELECT user0_.Id as Id2_0_, user0_.Name as Name2_0_, user0_.Street as Street2_0_, user0_.Country as Country2_0_, user0_.City as City2_0_, user0_.ZipCode as ZipCode2_0_, user0_.HouseNumber as HouseNum7_2_0_ FROM Users user0_ WHERE user0_.Id = 1 /* @p0 */ -- statement #3 SELECT currentlyr0_.[User] as User1_1_, currentlyr0_.Book as Book1_, book1_.Id as Id0_0_, book1_.Name as Name0_0_, book1_.ImageUrl as ImageUrl0_0_, book1_.Image as Image0_0_, book1_.Author as Author0_0_ FROM UsersReadingBooks currentlyr0_ left outer join Books book1_ on currentlyr0_.Book = book1_.Id WHERE currentlyr0_.[User] = 1 /* @p0 */ -- statement #4 SELECT queue0_.[User] as User1_1_, queue0_.Book as Book1_, queue0_.[Index] as Index3_1_, book1_.Id as Id0_0_, book1_.Name as Name0_0_, book1_.ImageUrl as ImageUrl0_0_, book1_.Image as Image0_0_, book1_.Author as Author0_0_ FROM UsersWaitingBooks queue0_ left outer join Books book1_ on queue0_.Book = book1_.Id WHERE queue0_.[User] = 1 /* @p0 */ -- statement #5 SELECT recommenda0_.[User] as User1_1_, recommenda0_.Book as Book1_, recommenda0_.[Index] as Index3_1_, book1_.Id as Id0_0_, book1_.Name as Name0_0_, book1_.ImageUrl as ImageUrl0_0_, book1_.Image as Image0_0_, book1_.Author as Author0_0_ FROM UsersRecommendedBooks recommenda0_ left outer join Books book1_ on recommenda0_.Book = book1_.Id WHERE recommenda0_.[User] = 1 /* @p0 */ -- statement #7 commit transaction
That seems strange, can you figure out why?
Bonus points for figuring out whatever it would be worth it to do the eager load optimization or not.
Comments
Looks like RSB handles message batches in a single transaction. Is it a bug or feature? BTW, serializable isolation level is almost guaranteed to cause problems in real world application.
Rafal,
Yes, that IS a feature.
And serializable isolation is actually a good thing in some circumstances. For one, it make it really simple to think about concurrency, and if all our data is user scoped, we aren't going to deal with a lot of complexity either.
Ah, I thought that sending an array of message is there only for performance reasons or developer convenience, didn't know about this little side effect.
Uhmm,
If the session crunching the messages is the same for all the messages in this example, then the first level cache is already an optimization. Eagerly fetch more then one collection at a time may results in a cartesian product query, that is worst than executing the 3 query to load the collection when needed.
Since one session is used, calls to session.Get <user(message.UserId) will only hit the database once.
The first time this is done, the books, queues and collections are eagerly loaded (if specified, you can't tell from the queries), but in a seperate select statements. I would not set fetchmode to join because this will lead to cartesian product. Further optimization could be to use a multicriteriaquery to send the three queries at once. But that would be an MS SQL specific optimization.
may be grouping all those queries into the same transaction might cause some more deadlocks?
Simone,
Can you envision a way in which two concurrent transactions (even for the same user) can deadlock with this code?
The problem in fact is with the type of transaction you're using
A transaction with Serializable Isolation places a range lock on the data set, preventing other users from updating or inserting rows into the data set until the transaction is complete.
Grouping those 4 queries on the same transaction will eventually put some more stress to the database if used in conjunction with a serializable transaction .
here I wrote a little bit more :)
http://smnbss.spaces.live.com/blog/cns!A117AA5E007A0648!1894.entry?&c02vws=1
Simone,
All the data that I am accessing is local to the current user
Sure, in your specific case it might not be a problem, but if you have other services accessing the same data with other queries it might create some problems. BTW, what is the problem then with that query? :)