[Date Prev][Date Next] [Chronological] [Thread] [Top]

(ITS#7473) back-mdb is 50x slower than back-hdb on datasets with numerous subtrees

Full_Name: Quanah Gibson-Mount
Version: RE24 12/11/2012
OS: Linux 2.6
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (74.196.25.250)


When testing on a database with over 40,000 subtrees, I discovered that search
results when using subtrees as a base for the search are at least 50x slower
with back-mdb than they are with back-hdb.  Details have been provided to
Howard, including the example dataset