Nice! Do you know why that benchmark show almost no read improvement but massive write improvement? The slides here (http://symas.com/mdb/20120322-UKUUG-MDB.pdf) suggest the opposite? Or am I reading them incorrectly?
As I understand it, that test used something like 32 server threads on a machine with only 4 cores. The UKUUG prezo used 16 server threads on a machine with 16 cores, which allowed all of the reads to run concurrently. The smaller number of CPU cores in the Zimbra test naturally reduces the concurrency advantage of LMDB.
Comments
Nice! Do you know why that benchmark show almost no read improvement but massive write improvement? The slides here (http://symas.com/mdb/20120322-UKUUG-MDB.pdf) suggest the opposite? Or am I reading them incorrectly?
As I understand it, that test used something like 32 server threads on a machine with only 4 cores. The UKUUG prezo used 16 server threads on a machine with 16 cores, which allowed all of the reads to run concurrently. The smaller number of CPU cores in the Zimbra test naturally reduces the concurrency advantage of LMDB.