我想这已经用不同的方式说过了。但是当你在处理一个处理器密集型算法时，你应该以牺牲其他所有东西为代价来简化最内部循环中的所有东西。

That may seem obvious to some, but it's something I try to focus on regardless of the language I'm working with. If you're dealing with nested loops, for example, and you find an opportunity to take some code down a level, you can in some cases drastically speed up your code. As another example, there are the little things to think about like working with integers instead of floating point variables whenever you can, and using multiplication instead of division whenever you can. Again, these are things that should be considered for your most inner loop.

有时，您可能会发现在内循环中对整数执行数学运算的好处，然后将其缩小为随后可以使用的浮点变量。这是一个牺牲一个部分的速度来提高另一个部分的速度的例子，但在某些情况下，这样做是值得的。

减少可变大小(在嵌入式系统中)

如果您的变量大小大于特定体系结构上的单词大小，则会对代码大小和速度产生重大影响。例如，如果你有一个16位系统，经常使用一个长int变量，然后意识到它永远不能超出范围(−32.768…32.767)考虑将其减少到短int。

从我的个人经验来看，如果一个程序已经准备好或几乎准备好了，但是我们意识到它占用了目标硬件程序内存的110%或120%，那么对变量进行快速归一化通常可以解决这个问题。

到这个时候，优化算法或部分代码本身可能会变得令人沮丧的徒劳:

重新组织整个结构，程序就不再像预期的那样工作，或者至少引入了许多错误。 做一些聪明的技巧:通常你花了很多时间优化一些东西，并发现代码大小没有或很小的减少，因为编译器无论如何都会优化它。

Many people make the mistake of having variables which exactly store the numerical value of a unit they use the variable for: for example, their variable time stores the exact number of milliseconds, even if only time steps of say 50 ms are relevant. Maybe if your variable represented 50 ms for each increment of one, you would be able to fit into a variable smaller or equal to the word size. On an 8 bit system, for example, even a simple addition of two 32-bit variables generates a fair amount of code, especially if you are low on registers, while 8 bit additions are both small and fast.

我大半辈子都在这里度过。大致的方法是运行你的分析器并记录它:

Cache misses. Data cache is the #1 source of stalls in most programs. Improve cache hit rate by reorganizing offending data structures to have better locality; pack structures and numerical types down to eliminate wasted bytes (and therefore wasted cache fetches); prefetch data wherever possible to reduce stalls. Load-hit-stores. Compiler assumptions about pointer aliasing, and cases where data is moved between disconnected register sets via memory, can cause a certain pathological behavior that causes the entire CPU pipeline to clear on a load op. Find places where floats, vectors, and ints are being cast to one another and eliminate them. Use __restrict liberally to promise the compiler about aliasing. Microcoded operations. Most processors have some operations that cannot be pipelined, but instead run a tiny subroutine stored in ROM. Examples on the PowerPC are integer multiply, divide, and shift-by-variable-amount. The problem is that the entire pipeline stops dead while this operation is executing. Try to eliminate use of these operations or at least break them down into their constituent pipelined ops so you can get the benefit of superscalar dispatch on whatever the rest of your program is doing. Branch mispredicts. These too empty the pipeline. Find cases where the CPU is spending a lot of time refilling the pipe after a branch, and use branch hinting if available to get it to predict correctly more often. Or better yet, replace branches with conditional-moves wherever possible, especially after floating point operations because their pipe is usually deeper and reading the condition flags after fcmp can cause a stall. Sequential floating-point ops. Make these SIMD.

我还喜欢做一件事:

将编译器设置为输出程序集清单，并查看它为代码中的热点函数发出了什么。所有那些聪明的优化，“一个好的编译器应该能够自动为你做”?实际的编译器可能不会执行这些操作。我见过GCC发出真正的WTF代码。

你知道吗，一根CAT6电缆能够比缺省的Cat5e UTP电缆更好地屏蔽外部干扰10倍?

对于任何非离线项目，尽管拥有最好的软件和硬件，但如果你的throughoutput很弱，那么这条细线就会挤压数据并给你带来延迟，尽管只有几毫秒……

此外，CAT6电缆的最大吞吐量更高，因为您实际上更有可能收到铜芯电缆，而不是CCA，铜芯包覆铝，这通常出现在所有标准CAT5e电缆中。

如果您面临丢包，丢包，那么提高24/7操作的吞吐量可靠性可以使您所寻找的不同。

对于那些追求家庭/办公室连接可靠性的人来说(并且愿意对今年的快餐店说不，在年底你可以在那里)，以知名品牌的CAT7电缆的形式为自己提供LAN连接的顶峰。

向它扔更多的硬件!

如果更好的硬件是一个选择，那么一定要去做。否则

Check you are using the best compiler and linker options. If hotspot routine in different library to frequent caller, consider moving or cloning it to the callers module. Eliminates some of the call overhead and may improve cache hits (cf how AIX links strcpy() statically into separately linked shared objects). This could of course decrease cache hits also, which is why one measure. See if there is any possibility of using a specialized version of the hotspot routine. Downside is more than one version to maintain. Look at the assembler. If you think it could be better, consider why the compiler did not figure this out, and how you could help the compiler. Consider: are you really using the best algorithm? Is it the best algorithm for your input size?