Re: [HACKERS] [PERFORM] how to plan for vacuum?

2007-01-26 Thread Jim Nasby
On Jan 25, 2007, at 10:33 AM, Ray Stell wrote: On Thu, Jan 25, 2007 at 08:04:49AM -0800, Joshua D. Drake wrote: It really depends on the system. Most of our systems run anywhere from 10-25ms. I find that any more than that, Vacuum takes too long. How do you measure the impact of setting i

Re: [HACKERS] [PERFORM] how to plan for vacuum?

2007-01-25 Thread Ray Stell
On Thu, Jan 25, 2007 at 08:04:49AM -0800, Joshua D. Drake wrote: > > It really depends on the system. Most of our systems run anywhere from > 10-25ms. I find that any more than that, Vacuum takes too long. How do you measure the impact of setting it to 12 as opposed to 15? -

Re: [HACKERS] [PERFORM] how to plan for vacuum?

2007-01-25 Thread Joshua D. Drake
Alvaro Herrera wrote: > Jim C. Nasby wrote: > >> I'll generally start with a cost delay of 20ms and adjust based on IO >> utilization. > > I've been considering set a default autovacuum cost delay to 10ms; does > this sound reasonable? It really depends on the system. Most of our systems run any

Re: [PERFORM] how to plan for vacuum?

2007-01-25 Thread Jim C. Nasby
Please cc the list so others can reply as well... On Thu, Jan 25, 2007 at 08:45:50AM +0100, Tomas Vondra wrote: > > On Wed, Jan 24, 2007 at 02:37:44PM +0900, Galy Lee wrote: > >> 1. How do we know if autovacuum is enough for my application, or should > >> I setup a vacuum manually from cron f

Re: [PERFORM] how to plan for vacuum?

2007-01-25 Thread Ray Stell
On Thu, Jan 25, 2007 at 07:29:20PM +0900, Galy Lee wrote: > so what is the principle to set them? > - keep dead space lower than some disk limit > - or keep the garbage rate lower than fillfactor > or any other general principle? How do you measure "dead space" and "garbage rate?" I'm a newbe

Re: [PERFORM] how to plan for vacuum?

2007-01-25 Thread Galy Lee
Jim C. Nasby wrote: On Wed, Jan 24, 2007 at 02:37:44PM +0900, Galy Lee wrote: 1. How do we know if autovacuum is enough for my application, or should I setup a vacuum manually from cron for my application? Generally I trust autovac unless there's some tables where it's critical that the

Re: [HACKERS] [PERFORM] how to plan for vacuum?

2007-01-24 Thread Jim C. Nasby
On Thu, Jan 25, 2007 at 12:52:02AM -0300, Alvaro Herrera wrote: > Jim C. Nasby wrote: > > > I'll generally start with a cost delay of 20ms and adjust based on IO > > utilization. > > I've been considering set a default autovacuum cost delay to 10ms; does > this sound reasonable? For a lightly lo

Re: [PERFORM] how to plan for vacuum?

2007-01-24 Thread Alvaro Herrera
Jim C. Nasby wrote: > I'll generally start with a cost delay of 20ms and adjust based on IO > utilization. I've been considering set a default autovacuum cost delay to 10ms; does this sound reasonable? -- Alvaro Herrerahttp://www.CommandPrompt.com/ PostgreSQL Rep

Re: [PERFORM] how to plan for vacuum?

2007-01-24 Thread Jim C. Nasby
On Wed, Jan 24, 2007 at 02:37:44PM +0900, Galy Lee wrote: > 1. How do we know if autovacuum is enough for my application, or should > I setup a vacuum manually from cron for my application? Generally I trust autovac unless there's some tables where it's critical that they be vacuumed frequent

Re: [PERFORM] how to plan for vacuum?

2007-01-23 Thread Galy Lee
Just have one example here: workload: run pgbench in 365x24x7 database size: 100GB the workload distribution: 06:00-24:00 100tps 00:00-06:00 20tps how should we plan vacuum for this situation to get the highest performance? Best regards Galy Galy Lee wrote: > Hi, > > For I can no

[PERFORM] how to plan for vacuum?

2007-01-23 Thread Galy Lee
Hi, For I can not find too much information about how to use vacuum, I want to ask some general information about the guideline of vacuum planning. 1. How do we know if autovacuum is enough for my application, or should I setup a vacuum manually from cron for my application? 2. How to set t