Dan*_*cik 2 c++ performance percentile
我的程序计算风险价值指标的蒙特卡罗模拟。为了尽可能简化,我有:
1/ simulated daily cashflows
2/ to get a sample of a possible 1-year cashflow,
I need to draw 365 random daily cashflows and sum them
Run Code Online (Sandbox Code Playgroud)
因此,每日现金流量是一个经验给定的分配函数,要采样 365 次。为此,我
1/ sort the daily cashflows into an array called *this->distro*
2/ calculate 365 percentiles corresponding to random probabilities
Run Code Online (Sandbox Code Playgroud)
我需要对年度现金流进行模拟,例如 10K 次,才能获得大量模拟年度现金流。准备好每日现金流量的分布函数,我进行抽样,例如...
for ( unsigned int idxSim = 0; idxSim < _g.xSimulationCount; idxSim++ )
{
generatedVal = 0.0;
for ( register unsigned int idxDay = 0; idxDay < 365; idxDay ++ )
{
prob = (FLT_TYPE)fastrand(); // prob [0,1]
dIdx = prob * dMaxDistroIndex; // scale prob to distro function size
// to get an index into distro array
_floor = ((FLT_TYPE)(long)dIdx); // fast version of floor
_ceil = _floor + 1.0f; // 'fast' ceil:)
iIdx1 = (unsigned int)( _floor );
iIdx2 = iIdx1 + 1;
// interpolation per se
generatedVal += this->distro[iIdx1]*(_ceil - dIdx );
generatedVal += this->distro[iIdx2]*(dIdx - _floor);
}
this->yearlyCashflows[idxSim] = generatedVal ;
}
Run Code Online (Sandbox Code Playgroud)
两个for循环内部的代码都进行线性插值。如果说 1000 美元对应于 prob=0.01,10000 美元对应于 prob=0.1,那么如果我没有 p=0.05 的经验数,我想通过插值获得 5000 美元。
问题:这段代码运行正确,尽管分析器说该程序在插值本身上花费了大约 60% 的运行时间。所以我的问题是,我怎样才能更快地完成这项任务?VTune 针对特定线路报告的示例运行时间如下:
prob = (FLT_TYPE)fastrand(); // 0.727s
dIdx = prob * dMaxDistroIndex; // 1.435s
_floor = ((FLT_TYPE)(long)dIdx); // 0.718s
_ceil = _floor + 1.0f; // -
iIdx1 = (unsigned int)( _floor ); // 4.949s
iIdx2 = iIdx1 + 1; // -
// interpolation per se
generatedVal += this->distro[iIdx1]*(_ceil - dIdx ); // -
generatedVal += this->distro[iIdx2]*(dIdx - _floor); // 12.704s
Run Code Online (Sandbox Code Playgroud)
破折号表示分析器不报告这些行的运行时。
任何提示将不胜感激。丹尼尔
编辑: c.fogelklou 和 MSalters 都指出了很大的改进。符合 c.fogelklou 所说的最好的代码是
converter = distroDimension / (FLT_TYPE)(RAND_MAX + 1)
for ( unsigned int idxSim = 0; idxSim < _g.xSimulationCount; idxSim++ )
{
generatedVal = 0.0;
for ( register unsigned int idxDay = 0; idxDay < 365; idxDay ++ )
{
dIdx = (FLT_TYPE)fastrand() * converter;
iIdx1 = (unsigned long)dIdx);
_floor = (FLT_TYPE)iIdx1;
generatedVal += this->distro[iIdx1] + this->diffs[iIdx1] *(dIdx - _floor);
}
}
Run Code Online (Sandbox Code Playgroud)
我沿着 MSalter 的路线所拥有的最好的是
normalizer = 1.0/(FLT_TYPE)(RAND_MAX + 1);
for ( unsigned int idxSim = 0; idxSim < _g.xSimulationCount; idxSim++ )
{
generatedVal = 0.0;
for ( register unsigned int idxDay = 0; idxDay < 365; idxDay ++ )
{
dIdx = (FLT_TYPE)fastrand()* normalizer ;
iIdx1 = fastrand() % _g.xDayCount;
generatedVal += this->distro[iIdx1];
generatedVal += this->diffs[iIdx1]*dIdx;
}
}
Run Code Online (Sandbox Code Playgroud)
第二个代码是大约。快 30%。现在,在 95 秒的总运行时间中,最后一行消耗了 68 秒。最后一行只消耗 3.2 秒,因此双倍乘法一定是魔鬼。我想到了 SSE - 将最后三个操作数保存到一个数组中,然后执行 this->diffs[i]*dIdx[i] 的向量乘法并将其添加到 this->distro[i] 但这段代码运行了 50%慢点。因此,我想我撞墙了。
非常感谢大家。D.
这是一个小优化的建议,消除了对 ceil、两次强制转换和一次乘法的需要。如果您在定点处理器上运行,这将解释为什么 float 和 int 之间的乘法和转换需要这么长时间。在这种情况下,如果 CPU 支持,请尝试使用定点优化或在编译器中打开浮点!
for ( unsigned int idxSim = 0; idxSim < _g.xSimulationCount; idxSim++ )
{
generatedVal = 0.0;
for ( register unsigned int idxDay = 0; idxDay < 365; idxDay ++ )
{
prob = (FLT_TYPE)fastrand(); // prob [0,1]
dIdx = prob * dMaxDistroIndex; // scale prob to distro function size
// to get an index into distro array
iIdx1 = (long)dIdx;
_floor = (FLT_TYPE)iIdx1; // fast version of floor
iIdx2 = iIdx1 + 1;
// interpolation per se
{
const FLT_TYPE diff = this->distro[iIdx2] - this->distro[iIdx1];
const FLT_TYPE interp = this->distro[iIdx1] + diff * (dIdx - _floor);
generatedVal += interp;
}
}
this->yearlyCashflows[idxSim] = generatedVal ;
}
Run Code Online (Sandbox Code Playgroud)