a couple other things: 1 lmdb is specifically designed for efficiency (optimised for read because it's for use in openldap). that *happens* to accidentally result in fast write performance under certain commonly-occurring [but *not* all] circumstances. 2 if you use compression on the values (i'd recommend snappy as it's a very fast partial-compression algorithm) whilst you gain on a reduced storage size you _will_ lose the speed benefits of not having that memcpy copy the value (like on every other key-value store), you would have to do decompress on every single value before you could access it.
l. -- Mailing list: https://launchpad.net/~mimblewimble Post to : mimblewimble@lists.launchpad.net Unsubscribe : https://launchpad.net/~mimblewimble More help : https://help.launchpad.net/ListHelp
