Devel::NYTProf v3 - 200908 (OUTDATED, see 201008)
0 likes1,055 views
The document discusses the evolution and features of various Perl profilers, particularly focusing on 'devel::nytprof' and its innovations in measuring time for subroutines and statements, offering detailed analysis for optimizing performance. It emphasizes the importance of profiling before optimization, providing a structured approach to identifying and fixing bottlenecks in code. The document concludes with insights on efficiency and warns against premature optimization while outlining methods for deeper and structural changes to improve performance.
![Avoid unpacking args
in very hot subs
sub foo { shift->delegate(@_) }
sub bar {
return shift->{bar} unless @_;
return $_[0]->{bar} = $_[1];
}](https://image.slidesharecdn.com/nytprof-200910-091112112914-phpapp02/85/Devel-NYTProf-v3-200908-OUTDATED-see-201008-56-320.jpg)
Devel::NYTProf v3 - 200908 (OUTDATED, see 201008)
- 1. Devel::NYTProf Perl Source Code Profiler Tim Bunce - October 2009
- 2. Devel::DProf • Oldest Perl Profiler —1995 • Design flaws make it practically useless on modern systems • Limited to 0.01 second resolution even for realtime measurements!
- 3. Devel::DProf Is Broken $ perl -we 'print "sub s$_ { sqrt(42) for 1..100 }; s$_({});n" for 1..1000' > x.pl $ perl -d:DProf x.pl $ dprofpp -r Total Elapsed Time = 0.108 Seconds Real Time = 0.108 Seconds Exclusive Times %Time ExclSec CumulS #Calls sec/call Csec/c Name 9.26 0.010 0.010 1 0.0100 0.0100 main::s76 9.26 0.010 0.010 1 0.0100 0.0100 main::s323 9.26 0.010 0.010 1 0.0100 0.0100 main::s626 9.26 0.010 0.010 1 0.0100 0.0100 main::s936 0.00 - -0.000 1 - - main::s77 0.00 - -0.000 1 - - main::s82
- 4. Lots of Perl Profilers • Take your pick... Devel::DProf | 1995 | Subroutine Devel::SmallProf | 1997 | Line Devel::AutoProfiler | 2002 | Subroutine Devel::Profiler | 2002 | Subroutine Devel::Profile | 2003 | Subroutine Devel::FastProf | 2005 | Line Devel::DProfLB | 2006 | Subroutine Devel::WxProf | 2008 | Subroutine Devel::Profit | 2008 | Line Devel::NYTProf | 2008 | Line & Subroutine
- 5. Evolution Devel::DProf | 1995 | Subroutine Devel::SmallProf | 1997 | Line Devel::AutoProfiler | 2002 | Subroutine Devel::Profiler | 2002 | Subroutine Devel::Profile | 2003 | Subroutine Devel::FastProf | 2005 | Line Devel::DProfLB | 2006 | Subroutine Devel::WxProf | 2008 | Subroutine Devel::Profit | 2008 | Line Devel::NYTProf v1 | 2008 | Line Devel::NYTProf v2 | 2008 | Line & Subroutine ...plus lots of innovations!
- 6. What To Measure? CPU Time Real Time Subroutines ? ? Statements ? ?
- 7. CPU Time vs Real Time • CPU time - Measures time CPU sent executing your code - Not (much) affected by other load on system - Doesn’t include time spent waiting for i/o etc. • Real time - Measures the elapsed time-of-day - Your time is affected by other load on system - Includes time spent waiting for i/o etc. CPU: only useful if high resolution Real: Most useful most of the time. Real users wait in real time! You want to know about slow I/O, slow db queries, delays due to thread contention, locks etc. etc
- 8. Sub vs Line • Subroutine Profiling - Measures time between subroutine entry and exit - That’s the Inclusive time. Exclusive by subtraction. - Reasonably fast, reasonably small data files • Problems - Can be confused by funky control flow - No insight into where time spent within large subs - Doesn’t measure code outside of a sub Funky: goto &sub, next/redo/last out of a sub, even exceptions
- 9. Sub vs Line • Line/Statement profiling - Measure time from start of one statement to next - Exclusive time (except includes built-ins & xsubs) - Fine grained detail • Problems - Very expensive in CPU & I/O - Assigns too much time to some statements - Too much detail for large subs (want time per sub) - Hard to get overall subroutine times
- 10. Devel::NYTProf
- 11. v1 Innovations • Fork by Adam Kaplan of Devel::FastProf - working at the New York Times • HTML report borrowed from Devel::Cover • More accurate: Discounts profiler overhead including cost of writing to the file • Test suite!
- 12. v2 Innovations • Profiles time per block! - Statement times can be aggregated to enclosing block and enclosing sub
- 13. v2 Innovations • Dual Profilers! - Is a statement profiler - and a subroutine profiler - At the same time
- 14. v2 Innovations • Subroutine profiler - tracks time per calling location - even for xsubs - calculates exclusive time on-the-fly - discounts cost of statement profiler - immune from funky control flow - in memory, writes to file at end - extremely fast
- 15. v2 Innovations • Statement profiler gives correct timing after leave ops - unlike previous statement profilers... - last statement in loops doesn’t accumulate time spent evaluating the condition - last statement in subs doesn’t accumulate time spent in remainder of calling statement Slightly dependent on perl version.
- 16. v2 Other Features • Profiles compile-time activity • Profiling can be enabled & disabled on the fly • Handles forks with no overhead • Correct timing for mod_perl • Sub-microsecond resolution • Multiple clocks, including high-res CPU time • Can snapshot source code & evals into profile • Built-in zip compression
- 18. Profiling Performance Time Size Perl x1 - DProf x 4.9 60,736KB SmallProf x 22.0 - FastProf x 6.3 42,927KB NYTProf x 3.9 11,174KB + blocks=0 x 3.5 9,628KB + stmts=0 x 2.5 205KB NYTProf v2.0 running perl 5.6.8 perlcritic 1.088 on lib/Perl/Critic/Policy
- 19. v3 Features • Profiles slow opcodes: system calls, regexps, ... • Subroutine caller name noted, for call-graph • Handles goto ⊂ e.g. AUTOLOAD • HTML report includes interactive TreeMaps • Outputs call-graph in Graphviz dot format
- 20. Running NYTProf perl -d:NYTProf ... perl -MDevel::NYTProf ... PERL5OPT=-d:NYTProf NYTPROF=file=/tmp/nytprof.out:addpid=1:slowops=1
- 21. Reporting: CSV • CSV - old, limited, dull $ nytprofcsv # Format: time,calls,time/call,code 0,0,0,sub foo { 0.000002,2,0.00001,print "in sub foon"; 0.000004,2,0.00002,bar(); 0,0,0,} 0,0,0,
- 22. Reporting: KCachegrind • KCachegrind call graph - new and cool - contributed by C. L. Kao. - requires KCachegrind $ nytprofcg # generates nytprof.callgraph $ kcachegrind # load the file via the gui
- 23. KCachegrind
- 24. Reporting: HTML • HTML report - page per source file, annotated with times and links - subroutine index table with sortable columns - interactive Treemaps of subroutine times - generates Graphviz dot file of call graph $ nytprofhtml # writes HTML report in ./nytprof/... $ nytprofhtml --file=/tmp/nytprof.out.793 --open
- 26. Summary Links to annotated source code Link to sortable table of all subs Timings for perl builtins
- 27. Exclusive vs. Inclusive • Exclusive Time = Bottom up - Detail of time spent “just here” - Where the time actually gets spent - Useful for localized (peephole) optimisation • Inclusive Time = Top down - Overview of time spent “in and below” - Useful to prioritize structural optimizations
- 29. Overall time spent in and below this sub (in + below) Color coding based on Median Average Deviation relative to rest of this file Timings for each location calling into, or out of, the subroutine
- 31. Treemap showing relative proportions of exclusive time Boxes represent subroutines Colors only used to show packages (and aren’t pretty yet) Hover over box to see details Click to drill-down one level in package hierarchy
- 32. Calls between packages Generates GraphViz files that can be used to produce these diagrams
- 34. Let’s take a look...
- 35. Optimizing Hints & Tips
- 36. Phase 0 Before you start
- 37. DONʼT DO IT! Don’t optimize until you have a real need to Concentrate on good design & implementation
- 38. “The First Rule of Program Optimization: Don't do it. The Second Rule of Program Optimization (for experts only!): Don't do it yet.” - Michael A. Jackson
- 39. Why not?
- 40. “More computing sins are committed in the name of efficiency (without necessarily achieving it) than for any other single reason - including blind stupidity.” - W.A. Wulf • programmer time $ > cpu time $. • likely to introduce bugs
- 41. “We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil. Yet we should not pass up our opportunities in that critical 3%.” - Donald Knuth
- 42. “We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil. Yet we should not pass up our opportunities in that critical 3%.” - Donald Knuth
- 43. How?
- 44. “Bottlenecks occur in surprising places, so don't try to second guess and put in a speed hack until you have proven that's where the bottleneck is.” - Rob Pike
- 45. “Measure twice, cut once.” - Old Proverb Measure Twice, cut once
- 46. Phase 1 Low Hanging Fruit
- 47. Low Hanging Fruit 1. Profile code running representative workload. 2. Look at Exclusive Time of subroutines. 3. Do they look reasonable? 4. Examine worst offenders. 5. Fix only simple local problems. 6. Profile again. 7. Fast enough? Then STOP! 8. Rinse and repeat once or twice, then move on. Bottlenecks move when changes are made.
- 48. “Simple Local Fixes” Changes unlikely to introduce bugs
- 49. Move invariant expressions out of loops
- 50. Avoid->repeated->chains ->of->accessors(...); Avoid->repeated->chains ->of->accessors(...); Use a temporary variable use a temporary variable
- 51. Use faster accessors Class::Accessor -> Class::Accessor::Fast --> Class::Accessor::Faster ---> Class::Accessor::Fast::XS
- 52. Avoid calling subs that don’t do anything! my $unused_variable = $self->get_foo; my $is_logging = $log->info(...); while (...) { $log->info(...) if $is_logging; ... }
- 53. Exit subs and loops early Delay initializations return if not ...a cheap test...; return if not ...a more expensive test...; my $foo = ...initializations...; ...body of subroutine...
- 54. Fix silly code - return exists $nav_type{$country}{$key} - ? $nav_type{$country}{$key} - : undef; + return $nav_type{$country}{$key};
- 55. Beware pathological regular expressions NYTPROF=slowops=2 NYTProf can now time regular expressions as if they’re subroutines
- 56. Avoid unpacking args in very hot subs sub foo { shift->delegate(@_) } sub bar { return shift->{bar} unless @_; return $_[0]->{bar} = $_[1]; }
- 57. Retest. Fast enough? STOP! Put the profiler down and walk away
- 59. Profile with a known workload E.g., 1000 identical requests So you can tell what’s ‘reasonable’.
- 60. Check Inclusive Times (especially top-level subs) Reasonable percentage for the workload? Proportion of phases: Read, modify, write. Extract, translate, load.
- 61. Check subroutine call counts Reasonable for the workload?
- 62. Add caching if appropriate to reduce calls Remember invalidation
- 63. Walk up call chain to find good spots for caching Remember invalidation
- 64. Creating many objects that don’t get used? Lightweight proxies e.g. DateTimeX::Lite
- 65. Retest. Fast enough? STOP! Put the profiler down and walk away
- 67. Push loops down - $object->walk($_) for @dogs; + $object->walk_these(@dogs); Often very effective. Especially if walk() has to do any initialisation.
- 68. Change the data structure hashes <–> arrays
- 69. Change the algorithm What’s the “Big O”? O(n 2) or O(logn) or ...
- 70. Rewrite hot-spots in C Inline::C
- 71. It all adds up! “I achieved my fast times by multitudes of 1% reductions” - Bill Raymond
- 72. Questions? Tim.Bunce@pobox.com @timbunce on twitter http://blog.timbunce.org