Add note that scaling factor should be >= # of clients.

diff --git a/contrib/pgbench/README.pgbench b/contrib/pgbench/README.pgbench
index 719eca639aa7f0a575658b27cb5383f29bcb46a1..a1c04f23d8b0c09484ecd12f50040101b2d5ebca 100644 (file)
--- a/contrib/pgbench/README.pgbench
+++ b/contrib/pgbench/README.pgbench
@@ -99,7 +99,9 @@ o options
                number of tuples generated will be multiple of the
                scaling factor. For example, -s 100 will imply 10M
                (10,000,000) tuples in the accounts table.
-               default is 1.
+               default is 1.  NOTE: scaling factor should be at least
+               as large as the largest number of clients you intend
+               to test; else you'll mostly be measuring update contention.
 
        -U login
                Specify db user's login name if it is different from