diff options
Diffstat (limited to 'system/doc/efficiency_guide')
| -rw-r--r-- | system/doc/efficiency_guide/profiling.xml | 171 |
1 files changed, 125 insertions, 46 deletions
diff --git a/system/doc/efficiency_guide/profiling.xml b/system/doc/efficiency_guide/profiling.xml index bf50a03fa6..f185456158 100644 --- a/system/doc/efficiency_guide/profiling.xml +++ b/system/doc/efficiency_guide/profiling.xml @@ -41,30 +41,87 @@ <p>Erlang/OTP contains several tools to help finding bottlenecks:</p> <list type="bulleted"> - <item><c>fprof</c> provides the most detailed information about - where the program time is spent, but it significantly slows down the - program it profiles.</item> - - <item><p><c>eprof</c> provides time information of each function - used in the program. No call graph is produced, but <c>eprof</c> has - considerable less impact on the program it profiles.</p> - <p>If the program is too large to be profiled by <c>fprof</c> or - <c>eprof</c>, the <c>cover</c> and <c>cprof</c> tools can be used - to locate code parts that are to be more thoroughly profiled using - <c>fprof</c> or <c>eprof</c>.</p></item> - - <item><c>cover</c> provides execution counts per line per - process, with less overhead than <c>fprof</c>. Execution counts - can, with some caution, be used to locate potential performance - bottlenecks.</item> - - <item><c>cprof</c> is the most lightweight tool, but it only - provides execution counts on a function basis (for all processes, - not per process).</item> + <item><p><seealso marker="tools:fprof"><c>fprof</c></seealso> provides + the most detailed information about where the program time is spent, + but it significantly slows down the program it profiles.</p></item> + + <item><p><seealso marker="tools:eprof"><c>eprof</c></seealso> provides + time information of each function used in the program. No call graph is + produced, but <c>eprof</c> has considerable less impact on the program it + profiles.</p> + <p>If the program is too large to be profiled by <c>fprof</c> or + <c>eprof</c>, <c>cprof</c> can be used to locate code parts that + are to be more thoroughly profiled using <c>fprof</c> or <c>eprof</c>.</p></item> + + <item><p><seealso marker="tools:cprof"><c>cprof</c></seealso> is the + most lightweight tool, but it only provides execution counts on a + function basis (for all processes, not per process).</p></item> + + <item><p><seealso marker="runtime_tools:dbg"><c>dbg</c></seealso> is the + generic erlang tracing frontend. By using the <c>timestamp</c> or + <c>cpu_timestamp</c> options it can be used to time how long function + calls in a live system take.</p></item> + + <item><p><seealso marker="tools:lcnt"><c>lcnt</c></seealso> is used + to find contention points in the Erlang Run-Time System's internal + locking mechanisms. It is useful when looking for bottlenecks in + interaction between process, port, ets tables and other entities + that can be run in parallel.</p></item> + </list> <p>The tools are further described in <seealso marker="#profiling_tools">Tools</seealso>.</p> + + <p>There are also several open source tools outside of Erlang/OTP + that can be used to help profiling. Some of them are:</p> + + <list type="bulleted"> + <item><url href="https://github.com/isacssouza/erlgrind">erlgrind</url> + can be used to visualize fprof data in kcachegrind.</item> + <item><url href="https://github.com/proger/eflame">eflame</url> + is an alternative to fprof that displays the profiling output as a flamegraph.</item> + <item><url href="https://ferd.github.io/recon/index.html">recon</url> + is a collection of Erlang profiling and debugging tools. + This tool comes with an accompanying E-book called + <url href="https://www.erlang-in-anger.com/">Erlang in Anger</url>.</item> + </list> + </section> + + <section> + <title>Memory profiling</title> + <pre>eheap_alloc: Cannot allocate 1234567890 bytes of memory (of type "heap").</pre> + <p>The above slogan is one of the more common reasons for Erlang to terminate. + For unknown reasons the Erlang Run-Time System failed to allocate memory to + use. When this happens a crash dump is generated that contains information + about the state of the system as it ran out of mmeory. Use the + <seealso marker="observer:cdv"><c>crashdump_viewer</c></seealso> to get a + view of the memory is being used. Look for processes with large heaps or + many messages, large ets tables, etc.</p> + <p>When looking at memory usage in a running system the most basic function + to get information from is <seealso marker="erts:erlang#memory/0"><c> + erlang:memory()</c></seealso>. It returns the current memory usage + of the system. <seealso marker="tools:instrument"><c>instrument(3)</c></seealso> + can be used to get a more detailed breakdown of where memory is used.</p> + <p>Processes, ports and ets tables can then be inspecting using their + respective info functions, i.e. + <seealso marker="erts:erlang#process_info_memory"><c>erlang:process_info/2 + </c></seealso>, + <seealso marker="erts:erlang#port_info_memory"><c>erlang:port_info/2 + </c></seealso> and + <seealso marker="stdlib:ets#info/1"><c>ets:info/1</c></seealso>. + </p> + <p>Sometimes the system can enter a state where the reported memory + from <c>erlang:memory(total)</c> is very different from the + memory reported by the OS. This can be because of internal + fragmentation within the Erlang Run-Time System. Data about + how memory is allocated can be retrieved using + <seealso marker="erts:erlang#system_info_allocator"> + <c>erlang:system_info(allocator)</c></seealso>. + The data you get from that function is very raw and not very plesant to read. + <url href="http://ferd.github.io/recon/recon_alloc.html">recon_alloc</url> + can be used to extract useful information from system_info + statistics counters.</p> </section> <section> @@ -80,6 +137,22 @@ tools on the whole system. Instead you want to concentrate on central processes and modules, which contribute for a big part of the execution.</p> + + <p>There are also some tools that can be used to get a view of the + whole system with more or less overhead.</p> + <list type="bulleted"> + <item><seealso marker="observer:observer"><c>observer</c></seealso> + is a GUI tool that can connect to remote nodes and display a + variety of information about the running system.</item> + <item><seealso marker="observer:etop"><c>etop</c></seealso> + is a command line tool that can connect to remote nodes and + display information similar to what the UNIX tool top shows.</item> + <item><seealso marker="runtime_tools:msacc"><c>msacc</c></seealso> + allows the user to get a view of what the Erlang Run-Time system + is spending its time doing. Has a very low overhead, which makes it + useful to run in heavily loaded systems to get some idea of where + to start doing more granular profiling.</item> + </list> </section> <section> @@ -128,7 +201,7 @@ performance impact. Using <c>fprof</c> is just a matter of calling a few library functions, see the <seealso marker="tools:fprof">fprof</seealso> manual page in - Tools .<c>fprof</c> was introduced in R8.</p> + Tools.</p> </section> <section> @@ -142,20 +215,6 @@ </section> <section> - <title>cover</title> - <p>The primary use of <c>cover</c> is coverage analysis to verify - test cases, making sure that all relevant code is covered. - <c>cover</c> counts how many times each executable line of code - is executed when a program is run, on a per module basis.</p> - <p>Clearly, this information can be used to determine what - code is run very frequently and can therefore be subject for - optimization. Using <c>cover</c> is just a matter of calling a - few library functions, see the - <seealso marker="tools:cover">cover</seealso> manual page in - Tools.</p> - </section> - - <section> <title>cprof</title> <p><c>cprof</c> is something in between <c>fprof</c> and <c>cover</c> regarding features. It counts how many times each @@ -202,16 +261,6 @@ <cell>No</cell> </row> <row> - <cell><c>cover</c></cell> - <cell>Per module to screen/file</cell> - <cell>Small</cell> - <cell>Moderate slowdown</cell> - <cell>Yes, per line</cell> - <cell>No</cell> - <cell>No</cell> - <cell>No</cell> - </row> - <row> <cell><c>cprof</c></cell> <cell>Per module to caller</cell> <cell>Small</cell> @@ -224,6 +273,37 @@ <tcaption>Tool Summary</tcaption> </table> </section> + + <section> + <title>dbg</title> + <p><c>dbg</c> is a generic Erlang trace tool. By using the + <c>timestamp</c> or <c>cpu_timestamp</c> options it can be used + as a precision instrument to profile how long time a function + call takes for a specific process. This can be very useful when + trying to understand where time is spent in a heavily loaded + system as it is possible to limit the scope of what is profiled + to be very small. + For more information, see the + <seealso marker="runtime_tools:dbg">dbg</seealso> manual page in + Runtime Tools.</p> + </section> + + <section> + <title>lcnt</title> + <p><c>lcnt</c> is used to profile interactions inbetween + entities that run in parallel. For example if you have + a process that all other processes in the system needs + to interact with (maybe it has some global configuration), + then <c>lcnt</c> can be used to figure out if the interaction + with that process is a problem.</p> + <p>In the Erlang Run-time System entities are only run in parallel + when there are multiple schedulers. Therefore <c>lcnt</c> will + show more contention points (and thus be more useful) on systems + using many schedulers on many cores.</p> + <p>For more information, see the + <seealso marker="tools:lcnt">lcnt</seealso> manual page in Tools.</p> + </section> + </section> <section> @@ -282,4 +362,3 @@ </list> </section> </chapter> - |