| From: | Tatsuo Ishii <t-ishii(at)sra(dot)co(dot)jp> |
|---|---|
| To: | Vadim Mikheev <vadim(at)krs(dot)ru> |
| Cc: | t-ishii(at)sra(dot)co(dot)jp, hackers(at)postgreSQL(dot)org |
| Subject: | Re: [HACKERS] READ COMMITTED isolevel is implemented ... |
| Date: | 1999-02-02 05:34:13 |
| Message-ID: | 199902020534.OAA13983@srapc451.sra.co.jp |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
>Tatsuo Ishii wrote:
>>
>> > and this is now the DEFAULT isolevel.
>>
>> But it seems that the standard says SERIALIZABLE is the default
>> isolation level (or at least the highest isolation level implemented
>> in the product), doesn't it?
>
>Yes, it does.
Then we should go for the standard way, I think.
>But Oracle, Informix, Sybase all use READ COMMITTED as default.
>Please decide youself - it doesn't matter much to me -:)
>I would like to see it 1. configure-able; 2. in pg_options;
>3. in command line args. I'll do this after beta started,
>if no one else before.
BTW, what is the advantage of READ COMMMITTED in PostgreSQL? I thought
the SERIALIZABLE should give us enough concurrency since we are using
MVCC. Could you give me some examples?
--
Tatsuo Ishii
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Hiroshi Inoue | 1999-02-02 06:27:42 | RE: [HACKERS] READ COMMITTED isolevel is implemented ... |
| Previous Message | Erich Stamberger | 1999-02-02 05:05:40 | trouble with rules |