Starting point: 1% Fragmentation
transaction type: TPC-B (sort of)
scaling factor: 100
number of clients: 50
number of transactions per client: 1000
number of transactions actually processed: 50000/50000
tps = 46.001927 (including connections establishing)
tps = 46.045008 (excluding connections establishing)
Total Fragmentation: 2%
File Fragmentation: 5%
transaction type: TPC-B (sort of)
scaling factor: 100
number of clients: 50
number of transactions per client: 1000
number of transactions actually processed: 50000/50000
tps = 50.507082 (including connections establishing)
tps = 50.558191 (excluding connections establishing)
Total Fragmentation: 2%
File Fragmentation: 5%
transaction type: TPC-B (sort of)
scaling factor: 100
number of clients: 100
number of transactions per client: 1000
number of transactions actually processed: 100000/100000
tps = 45.976461 (including connections establishing)
tps = 46.026986 (excluding connections establishing)
Total Fragmentation: 2%
File Fragmentation: 5%
Initialize with scaling of 100
Total Fragmentation: 2%
File Fragmentation: 5%
transaction type: TPC-B (sort of)
scaling factor: 100
number of clients: 50
number of transactions per client: 1000
number of transactions actually processed: 50000/50000
tps = 43.054164 (including connections establishing)
tps = 43.090163 (excluding connections establishing)
Total Fragmentation: 2%
File Fragmentation: 5%
O.k. my guess here is that autovacuum is keeping up with the actual
pgbench runs and thus causing reusable tuples? Does that make sense?
Joshua D. Drake
--
=== The PostgreSQL Company: Command Prompt, Inc. ===
Sales/Support: +1.503.667.4564 || 24x7/Emergency: +1.800.492.2240
Providing the most comprehensive PostgreSQL solutions since 1997 http://www.commandprompt.com/
Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate