Comments on: Misimproving performance problems with INSERT DELAYED http://code.openark.org/blog/mysql/misimproving-performance-problems-with-insert-delayed Blog by Shlomi Noach Wed, 01 Feb 2012 20:47:51 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: shlomi http://code.openark.org/blog/mysql/misimproving-performance-problems-with-insert-delayed/comment-page-1#comment-9394 shlomi Fri, 15 Jan 2010 06:58:40 +0000 http://code.openark.org/blog/?p=1745#comment-9394 Hi Morgan, Thanks for the clarification :) Hi Morgan,
Thanks for the clarification :)

]]> By: Morgan Tocker http://code.openark.org/blog/mysql/misimproving-performance-problems-with-insert-delayed/comment-page-1#comment-9392 Morgan Tocker Fri, 15 Jan 2010 06:09:42 +0000 http://code.openark.org/blog/?p=1745#comment-9392 Hi Shlomi, Great post. While I agree with you, I would clarify one thing: Performance is really the measure of two things; (a) response (b) capacity. INSERT delayed doesn't improve capacity in any way, but it can reduce the amount of time a user is sitting there waiting if the app is single threaded (like many web apps are). Having said that, I've never liked it. I've often thought of it as like a poor man's asynchronous queue. In many cases the right tool is probably a message queuing system designed for this problem like openmq/activemq/rabbitmq etc. Hi Shlomi,

Great post. While I agree with you, I would clarify one thing:

Performance is really the measure of two things; (a) response (b) capacity. INSERT delayed doesn't improve capacity in any way, but it can reduce the amount of time a user is sitting there waiting if the app is single threaded (like many web apps are). Having said that, I've never liked it. I've often thought of it as like a poor man's asynchronous queue.

In many cases the right tool is probably a message queuing system designed for this problem like openmq/activemq/rabbitmq etc.

]]>