Обсуждение: impact of auto explain on overall performance
Hello,
i’m currently working on a high Performance Database and want to make sure that whenever there are slow queries during regular operations i’ve got all Information about the query in my logs. So auto_explain come to mind, but the documentation explicitly states that it Comes at a cost. My Question is, how big is the latency added by auto_explain in percentage or ms ?
Best regards,
stephan
On Thu, Mar 14, 2019 at 07:29:17AM +0000, Stephan Schmidt wrote: > i’m currently working on a high Performance Database and want to make sure that whenever there are slow queries duringregular operations i’ve got all Information about the query in my logs. So auto_explain come to mind, but the documentationexplicitly states that it Comes at a cost. My Question is, how big is the latency added by auto_explain in percentageor ms ? https://www.postgresql.org/docs/current/auto-explain.html |log_analyze ... |When this parameter is on, per-plan-node timing occurs for all statements executed, whether or not they run long enoughto actually get logged. This can have an extremely negative impact on performance. Turning off auto_explain.log_timingameliorates the performance cost, at the price of obtaining less information. |auto_explain.log_timing (boolean) |auto_explain.log_timing controls whether per-node timing information is printed when an execution plan is logged; it's equivalentto the TIMING option of EXPLAIN. The overhead of repeatedly reading the system clock can slow down queries significantlyon some systems, so it may be useful to set this parameter to off when only actual row counts, and not exacttimes, are needed. This parameter has no effect unless auto_explain.log_analyze is enabled. This parameter is on bydefault. Only superusers can change this setting. I believe the cost actually varies significantly with the type of plan "node", with "nested loops" incurring much higher overhead. I think you could compare using explain(analyze) vs explain(analyze,timing off). While you're at it, compare without explain at all. I suspect the overhead is inconsequential if you set log_timing=off and set log_min_duration such that only the slowest queries are logged. Then, you can manually run "explain (analyze,costs on)" on any problematic queries to avoid interfering with production clients. Justin
On 3/14/19 9:23 AM, Justin Pryzby wrote: > On Thu, Mar 14, 2019 at 07:29:17AM +0000, Stephan Schmidt wrote: >> i’m currently working on a high Performance Database and want to make sure that whenever there are slow queries duringregular operations i’ve got all Information about the query in my logs. So auto_explain come to mind, but the documentationexplicitly states that it Comes at a cost. My Question is, how big is the latency added by auto_explain in percentageor ms ? > > https://www.postgresql.org/docs/current/auto-explain.html > |log_analyze > ... > |When this parameter is on, per-plan-node timing occurs for all statements executed, whether or not they run long enoughto actually get logged. This can have an extremely negative impact on performance. Turning off auto_explain.log_timingameliorates the performance cost, at the price of obtaining less information. > > |auto_explain.log_timing (boolean) > |auto_explain.log_timing controls whether per-node timing information is printed when an execution plan is logged; it'sequivalent to the TIMING option of EXPLAIN. The overhead of repeatedly reading the system clock can slow down queriessignificantly on some systems, so it may be useful to set this parameter to off when only actual row counts, and notexact times, are needed. This parameter has no effect unless auto_explain.log_analyze is enabled. This parameter is onby default. Only superusers can change this setting. > > I believe the cost actually varies significantly with the type of plan "node", > with "nested loops" incurring much higher overhead. > > I think you could compare using explain(analyze) vs explain(analyze,timing > off). While you're at it, compare without explain at all. > > I suspect the overhead is inconsequential if you set log_timing=off and set > log_min_duration such that only the slowest queries are logged. > > Then, you can manually run "explain (analyze,costs on)" on any problematic > queries to avoid interfering with production clients. > > Justin > You should also consider auto_explain.sample_rate: auto_explain.sample_rate causes auto_explain to only explain a fraction of the statements in each session. The default is 1, meaning explain all the queries. In case of nested statements, either all will be explained or none. Only superusers can change this setting. This option is available since 9.6 Regards
On Thu, Mar 14, 2019 at 3:29 AM Stephan Schmidt <schmidt@dltmail.de> wrote:
Hello,
i’m currently working on a high Performance Database and want to make sure that whenever there are slow queries during regular operations i’ve got all Information about the query in my logs. So auto_explain come to mind, but the documentation explicitly states that it Comes at a cost. My Question is, how big is the latency added by auto_explain in percentage or ms ?
You will have to measure it yourself and see. It depends on your hardware, OS, and OS version, and PostgreSQL version. And the nature of your queries. If you have auto_explain.log_timing=on, then I find that large sorts are the worst impacted. So if you have a lot of those, you should be careful.
On older kernels, I would run with auto_explain.log_timing=off. On newer kernels where you can read the clock from user-space, I run with auto_explain.log_timing=on. I find the slowdown noticeable with careful investigation (around 3%, last time I carefully investigated it), but usually well worth paying to have actual data to work with when I find slow queries in the log. I made a special role with auto_explain disabled for use with a few reporting queries with large sorts, both to circumvent the overhead and to avoid spamming the log with slow queries I already know about.
Cheers,
Jeff
On 3/14/19 00:29, Stephan Schmidt wrote: > i’m currently working on a high Performance Database and want to make > sure that whenever there are slow queries during regular operations i’ve > got all Information about the query in my logs. So auto_explain come to > mind, but the documentation explicitly states that it Comes at a cost. > My Question is, how big is the latency added by auto_explain in > percentage or ms ? One thought - what if the problem query is a 4ms query that just went to 6ms but it's executed millions of times per second? That would create a 150% increase to the load on the system. The approach I've had the most success with is to combine active session sampling (from pg_stat_activity) with pg_stat_statements (ideally with periodic snapshots) to identify problematic SQL statements, then use explain analyze after you've identified them. There are a handful of extensions on the internet that can do active session sampling for you, and I've seen a few scripts that can be put into a scheduler to capture snapshots of stats tables. Maybe something to consider in addition to the auto_explain stuff. -Jeremy -- http://about.me/jeremy_schneider