PostgreSQL 10 High Performance - Third Edition

Summary by usage count

That doesn't give any interesting information about the usage counts of either table though; this query will be shown to them:

SELECT 
  usagecount,count(*),isdirty 
FROM pg_buffercache 
GROUP BY isdirty,usagecount 
ORDER BY isdirty,usagecount; 

The results don't prove anything interesting so far, they just suggest there's not very much accumulating a high usage count on a straight percentage basis:

 usagecount | count | isdirty 
------------+-------+--------- 
          0 | 10544 | f 
          1 | 12096 | f 
          2 |  2709 | f 
          3 |  2308 | f 
          4 |  3119 | f 
          5 |  1992 | f

 usagecount | count | isdirty  
------------+-------+--------- 
          0 |  7177 | f 
          1 |  7948 | f 
          2 |   923 | f 
          3 |   168 | f 
          4 |    51 | f 
          5 |   117 | f