August 3, 2015 at 2:13 am
Last week I monitored the Page Life Expectancy on a system.
In a while loop with a build in delay of 1 minute I did a :
SELECT '--' [--], getdate() stamp, cntr_value AS 'Page Life Expectancy'
FROM sys.dm_os_performance_counters
WHERE object_name = 'SQLServer:Buffer Manager' AND counter_name = 'Page life expectancy'
And noticed an increase of 80 seconds of a PLE in one minute.
This was outside my expectation.
I know that the PLE is a prediction. But would like to know how it it establisched.
Ben
August 3, 2015 at 5:22 am
It's just based on how long the pages have stayed in memory. So from one minute to the next, a page has stayed in memory, at least 60 seconds, so that accounts for the majority of that change. If that same page had been in memory for another 20 seconds during the previous run of your query, it wouldn't have necessarily showed up, but some other page that had been in longer than 20, but less than 80 was in there. Chances are good, if you were seeing the same page being measured for duration that it incremented by 60 seconds each time after that.
Don't get hung too much on PLE. It's just a general health indicator and not a major one at that. It shows the volatility of your memory, but not whether that volatility is something to worry about. I'd look at other memory counters such as pages/sec, page faults/sec and memory grants pending. Also use the ring buffers to look for out of memory messages and focus on your wait statistics as indicators for where the bottlenecks are.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
August 3, 2015 at 7:12 am
Grant Fritchey (8/3/2015)
Don't get hung too much on PLE.
Thanks for you reply.
I do tend to get hung up to much by 'numbers' in general, and like to understand how these numbers get 'established'. I should not do this, but it is in my genetics.
Thanks for your advise.
Ben
August 3, 2015 at 7:57 am
ben.brugman (8/3/2015)
Grant Fritchey (8/3/2015)
Don't get hung too much on PLE.Thanks for you reply.
I do tend to get hung up to much by 'numbers' in general, and like to understand how these numbers get 'established'. I should not do this, but it is in my genetics.
Thanks for your advise.
Ben
Actually, such intellectual curiosity is a cherished quality in superiors, peers, and subordinates. Don't ever lose that quality.
--Jeff Moden
Change is inevitable... Change for the better is not.
August 3, 2015 at 8:13 am
Jeff Moden (8/3/2015)
ben.brugman (8/3/2015)
Grant Fritchey (8/3/2015)
Don't get hung too much on PLE.Thanks for you reply.
I do tend to get hung up to much by 'numbers' in general, and like to understand how these numbers get 'established'. I should not do this, but it is in my genetics.
Thanks for your advise.
Ben
Actually, such intellectual curiosity is a cherished quality in superiors, peers, and subordinates. Don't ever lose that quality.
I agree. I sure wasn't attempting to squish any inquiry or attempts at understanding.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
August 3, 2015 at 8:32 am
Viewing 6 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply