hi, Yes, I was really thinking about the InnoDB Tables, Thank you very much for such a systematic and accurate reply,
Thanks and regards, Chetan -----Original Message----- From: Jeremy Zawodny [mailto:[EMAIL PROTECTED]] Sent: Friday, January 25, 2002 1:10 PM To: Chetan Lavti Cc: [EMAIL PROTECTED] Subject: Re: Help needed !!! On Fri, Jan 25, 2002 at 01:05:12PM +0530, Chetan Lavti wrote: > > I am going to use the MySQL version 3.23.47 for as our database. > The issue is which table type to use. > > I want that the database should be (memory-resident). I have tried > with the MyISAM tables which doesn't solves my purpose as in this > case the data are stored in the files. Right. > I have also tried with the HEAP tables that uses a hashed index and > are stored in memory. This makes them very fast, but if MySQL > crashes I am loosing all data stored in that. As, I have created one > table as a heap type but when I make my server down and start it > again all rows created nowhere exists. Correct. It sounds like you might want to look at InnoDB tables. They are disk based, but the InnoDB table handler can use a significant amount of RAM (if you allow it to) to cache index *and* record data. MyISAM only caches index data in memory. The InnoDB buffer pool is where this cached data is stored. > So, if anybody can suggest me any process by which I can populated > HEAP table(es) from a duplicate table(es) (which is on secondary > storage) at the time of startup. Also, any runtime modifications in > the tables will need to be updated in both the tables (one in memory > and another one on disk). If any other method by which I can > achieve the same. There was talk last year of implementing a hybrid HEAP/MyISAM table type (originally motived by the DBA at Slashdot). Upon startup, the MyISAM table would be loaded into a RAM-based HEAP-table. Any changes to the HEAP table would get written to the underlying MyISAM table as well. Read-only queries, of course, would be run against the RAM-based HEAP table. I suspect that will never be implemented, since InnoDB and it's buffer pool go a long way toward solving the same problem. Hope that helps, Jeremy -- Jeremy D. Zawodny, <[EMAIL PROTECTED]> Technical Yahoo - Yahoo Finance Desk: (408) 349-7878 Fax: (408) 349-5454 Cell: (408) 685-5936 MySQL 3.23.41-max: up 22 days, processed 513,100,530 queries (265/sec. avg) --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
