Neil Padgett <[EMAIL PROTECTED]> writes: > Well. Currently the runs are the typical pg_bench runs. With what parameters? If you don't initialize the pg_bench database with "scale" proportional to the number of clients you intend to use, then you'll naturally get huge lock contention. For example, if you use scale=1, there's only one "branch" in the database. Since every transaction wants to update the branch's balance, every transaction has to write-lock that single row, and so everybody serializes on that one lock. Under these conditions it's not surprising to see lots of lock waits and lots of useless runs of the deadlock detector ... regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 4: Don't 'kill -9' the postmaster
- Re: [HACKERS] Spinlock performance improveme... Bruce Momjian
- Re: [HACKERS] Spinlock performance impro... Tom Lane
- Re: [HACKERS] Spinlock performance ... Bruce Momjian
- Re: [HACKERS] Spinlock performance improvement proposal Marc G. Fournier
- Re: [HACKERS] Spinlock performance improvement propo... Bruce Momjian
- Re: [HACKERS] Spinlock performance improvement proposal Tom Lane
- Re: [HACKERS] Spinlock performance improvement proposal Neil Padgett
- Re: [HACKERS] Spinlock performance improvement propo... Bruce Momjian
- Re: [HACKERS] Spinlock performance improvement proposal Tom Lane
- Re: [HACKERS] Spinlock performance improvement proposal Neil Padgett
- Re: [HACKERS] Spinlock performance improvement proposal Tom Lane
- Re: [HACKERS] Spinlock performance improvement proposal Ian Lance Taylor
- Re: [HACKERS] Spinlock performance improvement proposal Myron Scott
- Re: [HACKERS] Spinlock performance improvement proposal Doug McNaught
- Re: [HACKERS] Spinlock performance improvement proposal D. Hageman