…replication became CPU bound (single thread) instead of IO bound. You could celebrate but hey…. you need to check …increasing and there seems to be a reasonable number which would allow replication to perform better … a lot faster. I‘ve seen some people applying partition to basically all of …
Post: To pack or not to pack - MyISAM Key compression
May 13, 2006 By 4 Comments
…needs to be uncompressed for each of these Larger pages means slower index lookup as you need to scan half the page in average to…bound workload. If your load is disk bound there are more reasons to have keys packed so cache hit is better. Especially if you…
Post: Shard-Query adds parallelism to queries
November 15, 2010 By 18 Comments
… you have only a single server, but it has enough resources to answer queries in parallel, then it can be used to add parallelism to… for my workload, then this test would be CPU bound. What data did I use? I loaded 55GB of the same data used in… CPU bound at that point. If I needed more performance I could divide the data between two or more shards, or if possible, I could…
Post: How Percona does a MySQL Performance Audit
November 24, 2008 By 17 Comments
… I want to keep secrets from you. To the contrary, I‘d love to share it all with you….in the error log is interesting and needs to be investigated. You might find that there are …You cannot consider any one of these factors (schema, queries, indexing) in isolation, because they are tightly bound…
Post: Heikki Tuuri Innodb answers - Part I
October 26, 2007 By 19 Comments
…you favour database “self tuning” or prefer to expose tuning options to be available to the user. HT: I favor self tuning. PZ: I…bound and you‘re scaling out you can get better performance by using couple of MySQL Servers on single node. You can make them to…
Post: How simple answer are you looking for ?
April 11, 2007 By 5 Comments
… you start to explain it in details, a lot of variables add in – covering indexes, skewed distribution vs lookup frequency, IO bound or CPU bound workload which may be far too complicated. It is also interesting the broad answer would not be helpful – there… too much to deal with, so I‘m trying to keep the answers tailored towards customer needs omitting details which should not be important and…
Post: How to load large files safely into InnoDB with LOAD DATA INFILE
July 3, 2008 By 29 Comments
…to be capturing the bytes in per second. You can see how the curves are roughly logarithmic, which is what you…I‘ll try to collect more stats. It would also be interesting to try this on an I/O-bound…
Post: How much overhead is caused by on disk temporary tables
August 16, 2007 By 26 Comments
… one will be used depends on the allowed tmp_table_size and also by the data which needs to be put to …of time and when it has to perform them anyway which can cause IO bound load even with small …you may think there is no use for it. Looking at this number I found it to be too good to be true as I…
Comment: Query Profiling with MySQL: Bypassing caches
November 26, 2008 By 6 Comments
… uncached you need to clean these caches. for MyISAM you can do SET GLOBAL key_buffer_size=0 and when set it back to default. You… is likely to have something cached, dictionary information etc. I would have Slow Query log enabled with Percona Patches so you can see… – if you see from EXPLAIN query scans 1000000 rows you do not need it to be IO bound to optimize it you can just estimate how it would be 
Post: Using delayed JOIN to optimize count(*) and LIMIT queries
April 6, 2007 By 22 Comments
… you would see main (fact) table which contains search fields and dimension tables which contain more information about facts and which need to be joined to get query result. If you‘re executing count(*) queries for such result sets MySQL will perform the join even if you… set (0.00 sec) mysql> select count(*) from fact where i select count(*) from fact left join dim on val=id…
