Received: from sss.pgh.pa.us (unknown [192.204.191.242]) by postgresql.org (Postfix) with ESMTP id C3F50475C12 for ; Wed, 24 Jul 2002 14:21:06 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.12.5/8.12.5) with ESMTP id g6OIKeVk012809; Wed, 24 Jul 2002 14:21:00 -0400 (EDT) To: Elielson Fontanezi Cc: "'Daryl Herzmann'" , pgsql-sql , pgsql-general Subject: Re: RES: RES: [SQL] Queries not using Index In-reply-to: References: Comments: In-reply-to Elielson Fontanezi message dated "Wed, 24 Jul 2002 13:13:19 -0300" Date: Wed, 24 Jul 2002 14:20:40 -0400 Message-ID: <12808.1027534840@sss.pgh.pa.us> From: Tom Lane X-Archive-Number: 200207/1489 X-Sequence-Number: 27976 Elielson Fontanezi writes: > Hum... What such thing strange. > Hash algorithms should be better than BTREE and RTREE algorithms. Perhaps. The problem with Postgres' hash indexes is that no one has worked on the hash-index code since Berkeley days (except for one or two minor bugfixes, I think). The btree code has gotten a lot more care and attention, so it is now much better than the hash code. I'm not sure whether it's really worth anyone's time to try to bring the hash index code up to speed. It *might* be better than btree for certain limited applications, if it were equally well implemented. Or it might not. You'd have to invest a lot of work to find out, and might well discover that your work was wasted. regards, tom lane