Keiter: I think that’s a dead end in this discussion, to be honest.

Akande: Salle: was your comment referenced above about MYSQL?

Macandog: Scott0_: See http://dev.mysql.com/doc/refman/5.6/en/operator-precedence.html

Cowdery: So precedence definitely is not an order

Izsak: Well, it is the order of rules evaluated on the row

Izsak: It doesn’t imply that rows will come in any specific order

Izsak: So if you say “a=X or a=Y”, it can evaluate X,Y on either X,Y or Y,X or .

Swancutt: If I heard the word “order” in a database context, I would not ***ociate it with precedence

Swancutt: I’d ***ociate it with. ordering.

Danfield: One million qps – http://developers.slashdot.org/story/15/09/22/2145258/c***andra-rewritten-in-c-ten-times-faster

Beede: Or more relevant here – https://www.flamings****.com/blog/2014/06/03/1-million-sql-queries-per-second-mysql-5-7-on-power8/

Swancutt: Danfield: on a single server, kinda a bizarro benchmark?

Swancutt: Danfield: I don’t normally benchmark distributed databases, but when I do, I do it on a single server!

Losh: Benchmarketing to the max

Kimball: Cpu usage is high-ish. amazon rds web console shows me 3 connections. a commandline  mysql client running SHOW FULL PROCESSLISTg shows me only a single process running, which is the ‘show full processlist’ query. how else can I figure out what’s eating the cpu?

Dambach: Perf top -g -p $pidof mysqld !

Izsak: Dambach: would fail for me :

Dambach: Or the process isn’t mysqld

Dambach: You’re using centos5 in prod!

Izsak: 851363 789346 737848 454147 271499 222186 172119 108289

Dambach: Oh this is RDS, can you even SSH into those machines

Tamburrelli: I’m not clear if the ‘perf top’ was intended for me? but my database is a rds.

Ockimey: No I cannot ssh into it

Izsak: Cariaso: I’m sure amazon is mining bitcoins on your instance

Fitten: Ok, it seemed like a pure mysql issue, but I’ll try in ##aws

Vehrenkamp: My instances are fine, just the rds.

Dambach: Does the CPU usage have something like a %steal column

Zaniboni: If anything this seemed a bit like https://blog.mozilla.org/it/2012/06/30/mysql-and-the-leap-second-high-cpu-and-the-fix/

Furstenberg: Many metrics, but nothing akin to steal

Izsak: Cariaso: can’t you file a support ticket with amazon?

Izsak: Every time I need some item replaced or something

Izsak: They’re quite responsive

Lovig: I can, and I’d agree they are. but this seemed like something I should try to address myself first. so here I is.

Izsak: It essentially tells me that innodb purge ****s

Izsak: Cariaso: maybe it is innodb purge for you?!?

Dambach: Same except partitioned tables and/or performance_schema

Izsak: My #1 function is ut_delay is this bad?

Izsak: Hm, found something weird

Izsak: Maybe I should put this as #1 complaint

Izsak: And see if it gets fixed by 2020

Warwick: Domas: you got me poking around enough to discover that the mysql-error-running.log has quite a bit in it

Izsak: Will have to play with purge batch size

Gleeson: With messages like “InnoDB: Warning: a long semaphore wait:” and lots of details

Mariani: If that’s suggestive, holler otherwise I’m diving back into google

Sikora: Innodb_adaptive_hash_index=0 seems promising

Mcvaigh: Where is example mysql DB located?

Danfield: Rexwin_: http://dev.mysql.com/doc/sakila/en/

Lazano: Okay guys, I need a slight bit of advice.

Lievano: I have Table A and Table B. Table B uses A as a foreign key that I join on. I need to retrieve all of the IDs from table A where two conditions p***. The stumbling part is ‘where none of the related rows in table B have a timestamp newer than 3 hours ago’