Buffer Pool Throughput

Was watching a training vid by Paul Randall and stumbled on some of his Buffer Pool Throughput Queries.

First you find the total BufferPool size for your databases by running this query:


SELECT count(*)*8/1024)/1024 AS 'Data Cache Size(GB)',
 CASE database_id
 WHEN 32767 THEN 'RESOURCEDB'
 ELSE db_name(database_id)
 END AS 'DatabaseName'
FROM sys.dm_os_buffer_descriptors
GROUP BY db_name(database_id) ,database_id
ORDER BY 'Data Cache Size(GB)' DESC

88.69 GB was the value for one of my production database servers. Then you view your Page Life Expectancy counter to determine how quickly your buffer pool is being completely flushed:


SELECT [object_name],
[counter_name],
[cntr_value]
FROM sys.dm_os_performance_counters
WHERE [object_name] LIKE '%Manager%'
AND [counter_name] = 'Page life expectancy'

At the time of this article I was sitting at a PLE of 476,848 seconds, roughly 5.5 Days. Of course this number moves around but it gives a good indication of how much data is moving in/out of the buffer pool. I am flushing and refilling 88.69GB of data every 5.5 days.

Using Jonathan Kehayias’s PLE calculation of DataCacheSize / 4 * 300, I arrive at 26,609 seconds which is 7 hours or so. I will need to do some trending to see if that number holds true. But interesting calculations regardless.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s