-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Chris Anderson wrote: >> It's not surprising that shorter ids also made a similar difference. >> If we wanted to experiment more, it'd be interesting to see the >> difference between long and short sequential ids, as well as between >> short random and short sequential.
I'll do the requisite experiments this weekend trying to see what has the most effect on file size (length or randomness). The file size will be measured post compaction. I'm not too concerned about insert time and the smaller the database file size the more ends up in RAM cache anyway. Note that I'll be using CouchDB 0.10.0 as shipped on Ubuntu 9.10 64 bit. Please shout if that is not representative. Roger -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAktHa4YACgkQmOOfHg372QSg/QCgzqZZ1eYmuzfdNFmTyk9jfVyO P5cAoIsQo5Zd1w5MINnftJE5Dmzyng1w =xsqO -----END PGP SIGNATURE-----
