Microsoft MVP성태의 닷넷 이야기
글쓴 사람
정성태 (techsharer at outlook.com)
홈페이지
첨부 파일
(연관된 글이 1개 있습니다.)

C# - 재귀호출을 스택 자료구조와 반복문을 이용해 대체하는 방법을 Paralle.For와 함께?

아래의 글에 대해,

C# - 재귀호출을 스택 자료구조와 반복문을 이용해 대체하는 방법 
; https://www.sysnet.pe.kr/2/0/1599

Parallel.For를 이용한 재귀호출에도 사용이 가능하냐는 질문이 있었습니다. 예제로 보내온 소스코드는 다음과 같은데요.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;

namespace MonteCarlo
{
    class Program
    {
        static void Main(string[] args)
        {
            double dblSt = 0, dblEn = 0;

            Console.Write("Input the Start and End Value for Integration: ");
            string strBuff = Console.ReadLine();

            string[] strParams = strBuff.Split(' ');
            dblSt = double.Parse(strParams[0]);
            dblEn = double.Parse(strParams[1]);

            double dblInt = MonteCarloIntegral((x) =>
            {
                return (x + Math.Sin(x));
            }, dblSt, dblEn);
            Console.WriteLine("The Integration Result = {0}", dblInt);

            Console.ReadKey();
        }

        static double MonteCarloIntegral(Func<double, double> Func, double Init, double Fin)
        {
            double dblVolume = Fin - Init, dblIntg = 0;

            if (dblVolume <= 0.05) // Monte Carlo Integration: I = integral f(x) dx = (1/n) * Sum(i = 1 to n) f(x[i])
            {
                Random rndRand = new Random(100);
                long lngTrial = 10000 * ((long) ((Func(Fin) - Func(Init)) / dblVolume) + 1);

                for (long cnt = 0; cnt < lngTrial; cnt++) // Using I = integral f(x) dx = integral f(x(t)) * dx(t)/dt dt
                {
                    double dblRand = rndRand.NextDouble();
                    double dblReal = Math.Sqrt((dblRand * dblVolume * (Fin + Init)) + Math.Pow(Init, 2)); 
                    dblIntg += (Math.Pow(Fin, 2) - Math.Pow(Init, 2)) * Func(dblReal) / (2 * dblReal);
                }

                dblIntg /= lngTrial;
            }
            else
            {
                double[] dblPartSums = new double[8]; 
                double dblPartVol = dblVolume / dblPartSums.Length; 

                Parallel.For(0, dblPartSums.Length, (cnt) =>
                {
                    double dblStart = Init + (cnt * dblPartVol), dblEnd = Init + ((cnt + 1) * dblPartVol); 
                    dblPartSums[cnt] = MonteCarloIntegral(Func, dblStart, dblEnd);
                });

                dblIntg += dblPartSums.Sum();
            }

            return dblIntg;
        }
    }
}

정말 가능한지... 한번 해볼까요? ^^

우선, 예제 코드를 쉽게 이해할 수 있도록 병렬처리를 원래의 for 루프로 돌려놓겠습니다.

double[] dblPartSums = new double[8]; 
double dblPartVol = dblVolume / dblPartSums.Length; 

for (cnt i = 0; cnt < dblPartSums.Length; cnt ++)
{
    double dblStart = Init + (cnt * dblPartVol), dblEnd = Init + ((cnt + 1) * dblPartVol); 
    dblPartSums[cnt] = MonteCarloIntegral(Func, dblStart, dblEnd);
}

dblIntg += dblPartSums.Sum();

그럼, 이제 그다지 어렵지 않게 Stack + 반복문 유형으로 바꿀 수 있습니다.

public struct SnapShotStruct
{
    public double Init;
    public double Fin;

    public double dblVolume;
    public double dblIntg;

    public int stage;
}

static double MonteCarloIntegralLoop(Func<double, double> Func, double Init, double Fin)
{
    double retVal = 0.0;

    Stack<SnapShotStruct> snapshotStack = new Stack<SnapShotStruct>();

    SnapShotStruct currentSnapshot = new SnapShotStruct();
    currentSnapshot.Init = Init;
    currentSnapshot.Fin = Fin;

    currentSnapshot.dblVolume = 0;
    currentSnapshot.dblIntg = 0;

    currentSnapshot.stage = 0;

    snapshotStack.Push(currentSnapshot);

    while (snapshotStack.Count != 0)
    {
        currentSnapshot = snapshotStack.Pop();

        switch (currentSnapshot.stage)
        {
            case 0:
                currentSnapshot.dblVolume = currentSnapshot.Fin - currentSnapshot.Init;

                if (currentSnapshot.dblVolume <= 0.05)
                {
                    Random rndRand = new Random(100);
                    long lngTrial = 10000 * ((long)((Func(currentSnapshot.Fin) - Func(currentSnapshot.Init)) / currentSnapshot.dblVolume) + 1);

                    for (long cnt = 0; cnt < lngTrial; cnt++)
                    {
                        double dblRand = rndRand.NextDouble();
                        double dblReal = Math.Sqrt((dblRand * currentSnapshot.dblVolume * (currentSnapshot.Fin + currentSnapshot.Init)) + Math.Pow(currentSnapshot.Init, 2));
                        currentSnapshot.dblIntg += (Math.Pow(currentSnapshot.Fin, 2) - Math.Pow(currentSnapshot.Init, 2)) * Func(dblReal) / (2 * dblReal);
                    }

                    currentSnapshot.dblIntg /= lngTrial;

                    retVal += currentSnapshot.dblIntg;
                    continue;
                }

                currentSnapshot.stage = 1;
                snapshotStack.Push(currentSnapshot);

                int length = 8;
                double dblPartVol = currentSnapshot.dblVolume / length;

                for (int i = 0; i < length; i++)
                {
                    double dblStrt = currentSnapshot.Init + (i * dblPartVol);
                    double dblEnd = currentSnapshot.Init + ((i + 1) * dblPartVol);

                    SnapShotStruct newSnapShot = new SnapShotStruct();
                    newSnapShot.Init = dblStrt;
                    newSnapShot.Fin = dblEnd;
                    newSnapShot.stage = 0;

                    snapshotStack.Push(newSnapShot);
                }
                break;

            case 1:
                retVal = currentSnapshot.dblIntg + retVal;
                continue;
        }
    }

    return retVal;
}

Stack + 반복문으로 구현된 이 코드를 이제 원래의 Parallel.For로 돌려보면 어떻게 될까요?

int length = 8;
double dblPartVol = currentSnapshot.dblVolume / length;

Parallel.For(0, length, (idx) =>
{
    double dblStrt = currentSnapshot.Init + (idx * dblPartVol);
    double dblEnd = currentSnapshot.Init + ((idx + 1) * dblPartVol);

    SnapShotStruct newSnapShot = new SnapShotStruct();
    newSnapShot.Init = dblStrt;
    newSnapShot.Fin = dblEnd;
    newSnapShot.stage = 0;

    snapshotStack.Push(newSnapShot);
});

별 쓸모가 없죠? ^^ 물론 이렇게 바꾼다고 해서 해답이 달라지는 것은 아닙니다. 하지만, 재귀호출을 Stack + 반복문으로 고치는 것 자체가 함수 호출을 리스트 형식으로 직렬화하는 것과 유사하기 때문에 Parallel.For를 호출하는 것도 역시 목록에 포함시켜 다음에 실행하도록 스케쥴링하는 것에 불과하므로 그다지 효용성이 없습니다.

만약 위의 코드를 굳이 병렬화 되도록 고쳐야 한다면 Parallel.For 내의 코드를 별도의 함수 처리로 바꿔야 하고 이는 다시 메서드의 재귀호출로 이어지기 때문에 결국 바꾸기 전과 후의 차이가 없게 됩니다.

그나저나, 이번 질문을 통해서 새롭게 알게 되는군요. 재귀호출이 loop 내에서 불려지는 경우, 재귀호출 방법으로는 병렬화가 가능하지만 Stack + 반복문으로 처리하는 경우 병렬화 처리가 무의미하다는 것!

(위의 처리 코드는 첨부 파일에 넣어두었으니 참고하세요.)




[이 글에 대해서 여러분들과 의견을 공유하고 싶습니다. 틀리거나 미흡한 부분 또는 의문 사항이 있으시면 언제든 댓글 남겨주십시오.]

[연관 글]






[최초 등록일: ]
[최종 수정일: 6/23/2021]

Creative Commons License
이 저작물은 크리에이티브 커먼즈 코리아 저작자표시-비영리-변경금지 2.0 대한민국 라이센스에 따라 이용하실 수 있습니다.
by SeongTae Jeong, mailto:techsharer at outlook.com

비밀번호

댓글 작성자
 



2014-01-20 04시07분
[자유주의] 오호~^^ 루프 내에서의 스택 + 반복문 처리는 병렬화가 무의미하군요. 저 코드를 비재귀적으로 바꿀 수 있을까 하고 고민을 했는데, 비로소 답을 알게 되었네요.^^ 정말로 감사합니다.^^
[guest]

... 91  92  93  94  95  96  97  98  [99]  100  101  102  103  104  105  ...
NoWriterDateCnt.TitleFile(s)
11494정성태4/16/201816343개발 환경 구성: 362. Azure Web Apps(App Services)에 사용자 DNS를 지정하는 방법
11493정성태4/16/201818110개발 환경 구성: 361. Azure Web App(App Service)의 HTTP/2 프로토콜 지원
11492정성태4/13/201816126개발 환경 구성: 360. Azure Active Directory의 사용자 도메인 지정 방법
11491정성태4/13/201819322개발 환경 구성: 359. Azure 가상 머신에 Web Application을 배포하는 방법
11490정성태4/12/201818264.NET Framework: 739. .NET Framework 4.7.1의 새 기능 - Configuration builders [1]파일 다운로드1
11489정성태4/12/201816037오류 유형: 463. 윈도우 백업 오류 - a Volume Shadow Copy Service operation failed.
11488정성태4/12/201819740오류 유형: 462. Unhandled Exception in Managed Code Snap-in - FX:{811FD892-5EB4-4E73-A147-F1E079E36C4E}
11487정성태4/12/201817966디버깅 기술: 115. windbg - 닷넷 메모리 덤프에서 정적(static) 필드 값을 조사하는 방법
11486정성태4/11/201817066오류 유형: 461. Error MSB4064 The "ComputeOutputOnly" parameter is not supported by the "VsTsc" task
11485정성태4/11/201825023.NET Framework: 738. C# - Console 프로그램이 Ctrl+C 종료 시점을 감지하는 방법파일 다운로드1
11484정성태4/11/201826281.NET Framework: 737. C# - async를 Task 타입이 아닌 사용자 정의 타입에 적용하는 방법파일 다운로드1
11483정성태4/10/201829561개발 환경 구성: 358. "Let's Encrypt"에서 제공하는 무료 SSL 인증서를 IIS에 적용하는 방법 (2) [1]
11482정성태4/10/201821495VC++: 126. CUDA Core 수를 알아내는 방법
11481정성태4/10/201833849개발 환경 구성: 357. CUDA의 인덱싱 관련 용어 - blockIdx, threadIdx, blockDim, gridDim
11480정성태4/9/201823753.NET Framework: 736. C# - API를 사용해 Azure에 접근하는 방법 [2]파일 다운로드1
11479정성태4/9/201818585.NET Framework: 735. Azure - PowerShell로 Access control(IAM)에 새로운 계정 만드는 방법
11478정성태11/8/201921659디버깅 기술: 115. windbg - 덤프 파일로부터 PID와 환경변수 등의 정보를 구하는 방법 [1]
11477정성태4/8/201818707오류 유형: 460. windbg - sos 명령어 수행 시 c0000006 오류 발생
11476정성태4/8/201820169디버깅 기술: 114. windbg - !threads 출력 결과로부터 닷넷 관리 스레드(System.Threading.Thread) 객체를 구하는 방법
11475정성태3/28/201822759디버깅 기술: 113. windbg - Thread.Suspend 호출 시 응용 프로그램 hang 현상에 대한 덤프 분석
11474정성태3/27/201821037오류 유형: 459. xperf: error: TEST.Event: Invalid flags. (0x3ec).
11473정성태3/22/201825969.NET Framework: 734. C# - Thread.Suspend 호출 시 응용 프로그램 hang 현상파일 다운로드2
11472정성태3/22/201819604개발 환경 구성: 356. GTX 1070, GTX 960, GT 640M의 cudaGetDeviceProperties 출력 결과
11471정성태3/20/201822793VC++: 125. CUDA로 작성한 RGB2RGBA 성능 [1]파일 다운로드1
11470정성태3/20/201825603오류 유형: 458. Visual Studio - CUDA 프로젝트 빌드 시 오류 C1189, expression must have a constant value
11469정성태3/19/201818499오류 유형: 457. error MSB3103: Invalid Resx file. Could not load file or assembly 'System.Windows.Forms, ...' or one of its dependencies.
... 91  92  93  94  95  96  97  98  [99]  100  101  102  103  104  105  ...