Why does multithreaded memory allocate / deallocate an intensive application, doesn't scale with the number of threads?

Note:

Original title

Why doesn't DWScript multithreaded JSON parser scale with the number of threads?

was changed because this issue is not related to the processing of JSON data with DWScript. The problem is with the default memory manager in Delphi XE2 for XE7 (tested XE2 and trial XE7), but the problem is new for this type of application.


I have a Win32 / Win64 vcl multithreaded application that processes JSON data in Delphi XE2.

Each thread parses the JSON data using TdwsJSONValue.ParseString(sJSON)

DWScript, reads the values ​​using DWScript methods, and stores the result as records.

For testing purposes, I am processing the same JSON data in every thread.

In the thead single mode, processing N

seconds in the thread is performed to process the data. Increasing the number of threads to M

lineary (approximately M * N

) increases the time within a single thread required to process the same data.

As a result, there is no improvement in speed. The other parts of these applications (delivering JSON data, storing the results in the target environment) scale out as expected.

What could be the reason? Any ideas appreciated.

Additional Information:

  • Tested on Win7 / 32 and Win7 / 64, Win8 / 64 dual-core to 12-core (w / w-out HT) systems

  • DWScript was chosen as the fastest available (tested a bunch, including: Superobject, Delphi embedded). SO behaves similarly to the JSON block from DWS.

  • Below is a complete console application that illustrates the problem. To run it, we need the json sample data available here: https://www.dropbox.com/s/4iuv87ytpcdugk6/json1.zip?dl=0 This file contains the data json1.dat

    for the first stream. For streams up to 16, just copy json1.dat to json2.dat ... json16.dat.

    Program and data must be in the same folder. To run: convert.exe N, where N is the number of threads.

    The program writes the execution time to msecs to stout - it is spent on the stream, the data parsing time and the time it takes for the TdwsJSONValue object to be dropped (Destroy). The application is _dwsjvData.Destroy;

    not scalable.


program Convert;

{$APPTYPE CONSOLE}

{$R *.res}

uses
  System.SysUtils,
  System.Diagnostics,
  System.Classes,
  dwsJSON in 'dwsJSON.pas',
  dwsStrings in 'dwsStrings.pas',
  dwsUtils in 'dwsUtils.pas',
  dwsXPlatform in 'dwsXPlatform.pas';

type

  TWorkerThread = class (TThread)
  private
    _iUid:  Integer;
    _swWatch:  TStopwatch;
    _lRunning:  Boolean;

    _sFileJSonData:  String;
    _fJsonData:  TextFile;

  protected
    constructor Create (AUid: Integer);
    procedure Execute; override;

  published
    property Running: Boolean read _lRunning;

  end;

  TConverter = class (TObject)
  private
    _swWatch0, _swWatch1, _swWatch2:  TStopwatch;

    _dwsjvData:  TdwsJSONValue;

  protected
    constructor Create;
    destructor Destroy; override;

    function Calculate (AUid: Integer; AJSonData: String; var AParse, ADestroy: Integer): Integer;
  end;

const
  MAX_THREADS = 16;

var
  iHowMany:  Integer;
  athWorker:  array [1..MAX_THREADS] of Pointer;
  aiElapsed:  array [1..MAX_THREADS] of Integer;
  aiElapsedParse:  array [1..MAX_THREADS] of Integer;
  aiElapsedDestroy:  array [1..MAX_THREADS] of Integer;
  aiFares:  array [1..MAX_THREADS] of Integer;
  swWatchT, swWatchP:  TStopwatch;


constructor TWorkerThread.Create (AUid: Integer);
begin
  inherited Create (True);

  _iUid := AUid;
  _swWatch := TStopwatch.Create;
  _sFileJSonData := ExtractFilePath (ParamStr (0)) + 'json' + Trim (IntToStr (_iUid)) + '.dat';

  _lRunning := False;

  Suspended := False;
end;

procedure TWorkerThread.Execute;
var
  j:  Integer;
  sLine:  String;
  slLines:  TStringList;

  oS:  TConverter;
begin
  _lRunning := True;

  oS := TConverter.Create;

  slLines := TStringList.Create;
  System.AssignFile (_fJsonData, _sFileJSonData);
  System.Reset (_fJsonData);
  j := 0;
  repeat
    System.Readln (_fJsonData, sLine);
    slLines.Add (sLine);
    Inc (j);
  until (j = 50);
//  until (System.Eof (_fJsonData));
  System.Close (_fJsonData);

  Sleep (1000);

  _swWatch.Reset;
  _swWatch.Start;

  aiFares [_iUid] := 0;
  aiElapsedParse [_iUid] := 0;
  aiElapsedDestroy [_iUid] := 0;
  for j := 1 to slLines.Count do
    aiFares [_iUid] := aiFares [_iUid] + oS.Calculate (_iUid, slLines.Strings [j - 1], aiElapsedParse [_iUid], aiElapsedDestroy [_iUid]);

  _swWatch.Stop;

  slLines.Free;
  os.Destroy;

  aiElapsed [_iUid] := _swWatch.ElapsedMilliseconds;

  _lRunning := False;
end;

constructor TConverter.Create;
begin
  inherited Create;

  _swWatch0 := TStopwatch.Create;
  _swWatch1 := TStopwatch.Create;
  _swWatch2 := TStopwatch.Create;
end;

destructor TConverter.Destroy;
begin
  inherited;
end;

function TConverter.Calculate (AUid: Integer; AJSonData: String; var AParse, ADestroy: Integer): Integer;
var
  jFare, jTotalFares, iElapsedParse, iElapsedDestroy, iElapsedTotal:  Integer;
begin
  _swWatch0.Reset;
  _swWatch0.Start;

  _swWatch1.Reset;
  _swWatch1.Start;
  _dwsjvData := TdwsJSONValue.ParseString (AJSonData);
  _swWatch1.Stop;
  iElapsedParse := _swWatch1.ElapsedMilliseconds;

  if (_dwsjvData.ValueType = jvtArray) then
  begin
    _swWatch2.Reset;
    _swWatch2.Start;

    jTotalFares := _dwsjvData.ElementCount;
    for jFare := 0 to (jTotalFares - 1) do
      if (_dwsjvData.Elements [jFare].ValueType = jvtObject) then
      begin

        _swWatch1.Reset;
        _swWatch1.Start;

        _swWatch1.Stop;
      end;
  end;

  _swWatch1.Reset;
  _swWatch1.Start;
  _dwsjvData.Destroy;
  _swWatch1.Stop;
  iElapsedDestroy := _swWatch1.ElapsedMilliseconds;

  _swWatch0.Stop;
  iElapsedTotal := _swWatch0.ElapsedMilliseconds;

  Inc (AParse, iElapsedParse);
  Inc (ADestroy, iElapsedDestroy);

  result := jTotalFares;
end;

procedure MultithreadStart;
var
  j:  Integer;
begin
  for j := 1 to iHowMany do
    if (athWorker [j] = nil) then
    begin
      athWorker [j] := TWorkerThread.Create (j);

      TWorkerThread (athWorker [j]).FreeOnTerminate := False;
      TWorkerThread (athWorker [j]).Priority := tpNormal;
    end;
end;

procedure MultithreadStop;
var
  j:  Integer;
begin
  for j := 1 to MAX_THREADS do
    if (athWorker [j] <> nil) then
    begin
      TWorkerThread (athWorker [j]).Terminate;
      TWorkerThread (athWorker [j]).WaitFor;

      TWorkerThread (athWorker [j]).Free;
      athWorker [j] := nil;
    end;
end;

procedure Prologue;
var
  j:  Integer;
begin
  iHowMany := StrToInt (ParamStr (1));

  for j := 1 to MAX_THREADS do
    athWorker [j] := nil;

  swWatchT := TStopwatch.Create;
  swWatchT.Reset;

  swWatchP := TStopwatch.Create;
  swWatchP.Reset;
end;

procedure RunConvert;

  function __IsRunning: Boolean;
  var
    j:  Integer;
  begin
    result := False;
    for j := 1 to MAX_THREADS do
      result := result or ((athWorker [j] <> nil) and TWorkerThread (athWorker [j]).Running);
  end;

begin

  swWatchT.Start;

  MultithreadStart;

  Sleep (1000);
  while (__isRunning) do
    Sleep (500);

  MultithreadStop;

  swWatchT.Stop;
  Writeln (#13#10, 'Total time:', swWatchT.ElapsedMilliseconds);
end;

procedure Epilogue;
var
  j:  Integer;
begin
  for j := 1 to iHowMany do
    Writeln ( #13#10, 'Thread # ', j, '  tot.time:', aiElapsed [j], '  fares:', aiFares [j], '  tot.parse:', aiElapsedParse [j], '  tot.destroy:', aiElapsedDestroy [j]);

  Readln;
end;

begin
  try
    Prologue;
    RunConvert;
    Epilogue;

  except
    on E: Exception do
      Writeln (E.ClassName, ': ', E.Message);
  end;
end.

      

0


source to share


3 answers


The solution is the default Exchange Manager Delphi XE2 or XE7 with Intel® Threading Building Blocks Memory Manager. In the sample app, it scales ca. lineary with up to 16 threads when the application is 64 bits.

update: with assumption that number of threads running is less than number of cores

This has been tested on machines ranging from 2cores / 4ht to 12cores / 24ht running KVM, virtualized Windows 7 with 124GB RAM.

The interesting thing is virtualization of Win 7. Memory allocation and deallocation is 2x faster than native Win 7.



Conclusion: if you are doing a lot of 10kB-10MB block memory allocation / deallocation operations in multithreaded application threads (more than 4-8 threads), use only Intel memory manager.

@André: thanks for the hint pointing me in the right direction!

Here is a block with TBB memory manager taken for tests, it should appear as 1st item list in the main project .dpr file

unit TBBMem;

interface

function  ScalableGetMem  (ASize: NativeInt): Pointer; cdecl; external 'tbbmalloc' name 'scalable_malloc';
procedure ScalableFreeMem (APtr: Pointer); cdecl; external 'tbbmalloc' name 'scalable_free';
function  ScalableReAlloc (APtr: Pointer; Size: NativeInt): Pointer; cdecl; external 'tbbmalloc' name 'scalable_realloc';

implementation

Function TBBGetMem (ASize: Integer): Pointer;
begin
  result := ScalableGetMem (ASize);
end;

Function TBBFreeMem (APtr: Pointer): Integer;
begin
  ScalableFreeMem (APtr);
  result := 0;
end;

Function TBBReAllocMem (APtr: Pointer; ASize: Integer): Pointer;
begin
  result := ScalableRealloc (APtr, ASize);
end;

const
  TBBMemoryManager:  TMemoryManager = ( GetMem: TBBGetmem;
                                        FreeMem: TBBFreeMem;
                                        ReAllocMem:  TBBReAllocMem; );
var
  oldMemoryManager:  TMemoryManager;

initialization
  GetMemoryManager (oldMemoryManager);
  SetMemoryManager (TBBMemoryManager);

finalization
  SetMemoryManager (oldMemoryManager);

end.

      

+1


source


Have you tried my scalable memory manager? Since Delphi (with fastmm built in) doesn't scale very well with strings and other memory related stuff: https://scalemm.googlecode.com/files/ScaleMM_v2_4_1.zip



And you can also try both modes of my profiler's profiler to see which part is the bottleneck: https://code.google.com/p/asmprofiler/

+1


source


I ran (replay) the FastCode MM Challenge test and the results were not good for TBB (also an out of memory exception in the block downsize test).

In short: ScaleMM2 and Google TCmalloc are the fastest in this difficult test, Fastmm and ScaleMM2 use the least memory.

Average Speed Performance: (Scaled so that the winner = 100%)
  XE6         :   70,4
  TCmalloc    :   89,1
  ScaleMem2   :  100,0
  TBBMem      :   77,8

Average Memory Performance: (Scaled so that the winner = 100%)
  XE6         :  100,0
  TCmalloc    :   29,6
  ScaleMem2   :   75,6
  TBBMem      :   38,4

      

FastCode Challenge: https://code.google.com/p/scalemm/source/browse/#svn%2Ftrunk%2FChallenge
TBB 4.3: https://www.threadingbuildingblocks.org/download

+1


source







All Articles