Re: Datatypes and performance
От | Maksim Likharev |
---|---|
Тема | Re: Datatypes and performance |
Дата | |
Msg-id | 56510AAEF435D240958D1CE8C6B1770A016D2D90@mailc03.aurigin.com обсуждение исходный текст |
Ответ на | Datatypes and performance ("Jay O'Connor" <joconnor@cybermesa.com>) |
Ответы |
Re: Datatypes and performance
|
Список | pgsql-general |
Wasn't any sarcasm in my words, as I sad I am really glad to see real numbers behind the story about no performance difference for varchar and text. I do not have performance metrics for varchars, but I have benchmarks and my observations about performance in general. Just want to say, I haven't seen much other DBMS systems other then Microsoft SQL, and Postgres 7.3.x, so my performance observations based on those systems. Selects, do not have any problems in general equal or slightly slower that MS SQL. Inserts/Updates sometimes slow, sometimes a nightmare, in general painfully slow. I guess as long as Postgres retrieves data fast enough I satisfied, but for dynamic DBs ( lots of inserts/updates ) .... P.S do not want to offend somebody or something, just saying what I see. -----Original Message----- From: Alvaro Herrera [mailto:alvherre@dcc.uchile.cl] Sent: Friday, July 04, 2003 8:13 PM To: Maksim Likharev Cc: PostgreSQL List Subject: Re: [GENERAL] Datatypes and performance On Fri, Jul 04, 2003 at 06:57:04PM -0700, Maksim Likharev wrote: > Ok, then if I store 4K - 1byte in a varchar and after that 4K + 1 byte, > and again and again, you are saying me that my text will be jerking > around 2 tables? In any case the tuple will probably "jerk around" different pages (I don't think it's different whether the pages are from two tables or from the same one). I fail to understand how that is a different performance problem than for any other datatype. I'm sure you've done some benchmark to determine that the varchar handling is as slow as you appear to be saying? -- Alvaro Herrera (<alvherre[a]dcc.uchile.cl>) "Postgres is bloatware by design: it was built to house PhD theses." (Joey Hellerstein, SIGMOD annual conference 2002)
В списке pgsql-general по дате отправления: