From: "Martin J. Dürst" <duerst@...>
Date: 2016-03-17T17:59:24+09:00
Subject: [ruby-core:74407] Re: [Ruby trunk Feature#12142] Hash tables with open addressing

On 2016/03/17 16:50, SASADA Koichi wrote:
> On 2016/03/17 15:00, shyouhei@ruby-lang.org wrote:
>> About 64bit versus 32bit index: several developers discussed this on this month's developer meeting.  Consensus there was that we do not want to limit Hash size, while it is true that over 99% of hashes are smaller than 4G entries, and it is definitely a good idea to optimize to them.  We did not reach a consensus as to how should we achieve that, though.

Matz also at the beginning of the discussion of this issue said that 
currently, he would emphasize speed over size, so if e.g. the 32-bit 
version led to a 10% speed increase, going with 32-bit would be okay.

We also agreed that currently, a Hash with more than 4G elements is 
mostly theoretical, but that will change in the future. So it may be 
okay to stay with 32-bit if we can change it easily to 64-bit in the future.

Unsubscribe: <mailto:ruby-core-request@ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>