Re: Add free-behind capability for large sequential scans
От | Neil Padgett |
---|---|
Тема | Re: Add free-behind capability for large sequential scans |
Дата | |
Msg-id | 1013622284.1571.64.camel@shoebox обсуждение исходный текст |
Ответ на | Re: Add free-behind capability for large sequential scans (Amit Kumar Khare <skamit2000@yahoo.com>) |
Ответы |
Re: Add free-behind capability for large sequential scans
|
Список | pgsql-hackers |
On Wed, 2002-02-13 at 03:11, Amit Kumar Khare wrote: [clip] > The problem is same here(Please correct me if I am > wrong) what they talk of. They have recognized like > Professor Stonebraker certain Access patterns (like > clustered sequential, looping sequential etc.)in > Database Systems and recomend a "Composite page > replacement policy". But how the buffer manager will > know what policy has to be applied? They say "When a > file is opened, its associated locality set size and > REPLACEMENT POLICY are given to the buffer manager". This is indeed one possibility. However, the problem, as pointed out in [1] is that in multi-user situations, getting sane results from query execution analysis is hard. The real problem is -- how do you handle the interaction of multiple simultaneous queries with the buffer pool? This problem led for a search for a new approach, which in turn led to simpler algorithms, like LRU-K [1] and 2Q [2]. I'd much rather we pursue algorithms of this type. Neil [1] E.H. O'Neil, P.E. O'Neil, and G. Weikum. The LRU-K page replacement algorithm for database disk buffering. _In Proceeedings of the 1993 ACM Sigmod International Conference on Management of Data_, pages 297-306, 1993. [2] Theodore Johnson and Dennis Shasha. 2Q: A Low Overhead High Performace Buffer Amanagement Replacement algorithm. In _Proceedings of the 20th VLDB Conference_, pages 439-450, 1994. -- Neil Padgett Red Hat Canada Ltd. E-Mail: npadgett@redhat.com 2323 Yonge Street, Suite #300, Toronto, ON M4P 2C9
В списке pgsql-hackers по дате отправления: