Avoiding deadlocks in concurrent programming

Eloff eloff777 at yahoo.com
Thu Jun 23 08:39:14 CEST 2005


Thanks for all of the replies, I'm glad I posted here, you guys have
been very helpful.

>Obviously, I only know what you've told us about your data, but 20-100
>queries?  That doesn't sound right ... RDBMSes are well-
>studied and well-understood; they are also extremely powerful when used
>to their potential.

Well I doubt there'd be even as many as 20 unique queries in any one
action, but 20-100 queries executed is about average, and it can be
over ten thousand for one action, clearly painfull to any RDBM or
programmer.

And as Paul McGuire points out, RDBMs don't help avoid deadlocks, in
fact it makes them easier to create in my opinion.

Paul Rubin, I think you're right, a single lock is the approach I'm
going to take. I don't want to seem stupid, but I don't understand what
you're talking about with the Queues, isn't a Queue just a synchronized
sequence? If you've got the time and you don't mind, I'd love a more
detailed explanation.

Niel, thanks for the link, I read through the article.

>I try to pick up crumbs of knowledge from my co-workers, and one of the
>smarter ones gave me this rubric for testing for deadlocks.

That's true, I'm going to remember that one.

Thanks a lot guys, I'm leaving on a trip tomorrow so I won't be able to
reply to this thread again, but I will read any other posts on it when
I come back.

-Dan




More information about the Python-list mailing list