Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a table. Due to the nature of the website, in that table, if i have 400+ users or so, it will have in it inserted 100,000 new rows per hour.

If i have near 4000 then for 2 hours or so, i will have a minimum of 500,000 rows inserted. This means about 800-1000 inserts per second, probably even way more than that.

How do i prepare the table for such intense use? I am talking about quite the lot of inserts per second in the table and i don't want it to get locked or slow down.

The table engine is MyISAM, but it could be altered to accommodate any changes

share|improve this question

it all depends on your usage pattern - do you need high performance for lookups in this table? will lookups be evenly distributed across the whole table or just latest data? if selects will mostly hit recently inserted rows - consider partitioning - check this and that.

myisam almost + heavy write load sounds like a bad idea to me. go innodb. learn about tuning of that engine, use fast disks, have enough memory / buffer pool config to fit at least indices in ram.

check postings at mysqlperformanceblog - you'll find a lot of good advice there.

spend a lot of time on benchmarking - see how the system behaves under prolonged load [you might see drops in performance due to checkpointing - is it acceptable? or maybe you need to get 5.5 or xtradb to solve that problem?]

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.