| From: | Mark Mielke <mark(at)mark(dot)mielke(dot)cc> |
|---|---|
| To: | Michael Glaesemann <grzm(at)seespotcode(dot)net> |
| Cc: | Hannu Krosing <hannu(at)skype(dot)net>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Gregory Stark <stark(at)enterprisedb(dot)com>, Kenneth Marshall <ktm(at)rice(dot)edu>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Hash index todo list item |
| Date: | 2007-09-06 21:17:12 |
| Message-ID: | 46E06E58.1080708@mark.mielke.cc |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Michael Glaesemann wrote:
> On Sep 6, 2007, at 10:53 , Mark Mielke wrote:
>> I don't like the truncating hash suggestion because it limits the
>> ability of a hash code to uniquely identify a key.
> AIUI, a hash can't be used as a unique identifier: it always needs to
> be rechecked due to the chance of collisions. There might be other
> issues with truncation, but preventing hashes from being unique isn't
> one of them.
Of course - that's why I used the word "limit".
Hash works best, when the key is unique, however. A 32-bit hash will be
many powers of 2 more unique than a 8-bit hash.
Cheers,
mark
--
Mark Mielke <mark(at)mielke(dot)cc>
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Decibel! | 2007-09-06 22:50:41 | Re: Just-in-time Background Writer Patch+Test Results |
| Previous Message | Kevin Grittner | 2007-09-06 21:15:06 | Re: [FEATURE REQUEST] Streaming Onlinebackup (Maybe OFFTOPIC) |