In the blog “Faster Checked Addition in C#” we saw how to add numbers safely in C# without using the checked key work and without exceptions. This raised performance, since exceptions have quite a bit of overhead. In this blog, I’ll extend this idea to the data parallel SIMD/SSE instructions of Intel and AMD processors, […]
Read more "Checked SIMD/SSE Addition in C#"Several improvements to C# .Sum() were shown in the “Better C# .Sum() in Many Ways” blog and made available in the HPCsharp nuget package. In this blog, I’ll explore more ways to improve .Sum() to raise its capabilities to the next level in performance while not giving up accuracy. BigInteger Summation C# provides a BigInteger […]
Read more "Better C# .Sum() in More Ways"One of my pet peeves is not using hardware to its fullest ability, such as all of memory bandwidth available. Copying is one of the simplest operations, which is memory bandwidth bound, and should use most of memory bandwidth. In this blog, I’ll examine C# copy operations from List to Array, to see how fast […]
Read more "Faster List.ToArray() and Copying in C#"Adding Two Integers – Surprise! Adding two integers in C# can produce surprising results. For instance, int sum = 2147483647; // Int32.MaxValue == 0x7FFFFFFF sum = sum + 1; The resulting sum is expected to be 2147483648, but is -2147483648 instead. uint sum = 4294967295; // UInt32.MaxValue == 0xFFFFFFFF sum = sum + 1; The […]
Read more "Faster Checked Addition in C#"C# Linq provides a convenient way to add up all of the array elements. var arr = new int[] { 5, 7, 16, 3 }; int arrSum = arr.Sum(); This .Sum() supports some of the built-in numeric types: int, long, float, double, and decimal. It provides a nicely consistent and convenient way of adding up […]
Read more "Better C# .Sum() in Many Ways"JavaScript has two kinds of arrays: standard ones and typed ones. These two kinds of arrays have different capabilities. Standard arrays can grow, but typed arrays can not. Standard arrays can hold numbers, strings, objects, etc.. Typed arrays can only hold specific numeric data types: 8-bit, 16-bit, 32-bit and 64 bit unsigned and signed integers, […]
Read more "Faster JavaScript Sorting Using Typed Arrays"In this blog I’ll assemble variety of resources for code performance optimization – from videos to blog posts to StackOverflow topics to university class instructions to books. I’ll update this post from time to time. Videos Books Blogs Gallery of Processor Cache Effects
Read more "Code Performance Optimization Resources"LSD Radix Sort algorithm has been around for a long time with first computer usage in 1950’s and 1960’s to sort punch cards by performing several passes at the stack of cards. It has resurfaced in the last decade with GPUs having hundreds and even thousands of processor with internal local truly random access memories […]
Read more "Faster LSD Radix Sort"Computer Scientists develop algorithms for all kinds of tasks, such as sorting numbers, searching the web for information, showing web pages and interacting with them, and so on – thousands of different algorithms. Even a task such as sorting a bunch of numbers, has more than 20 different algorithms. When you need to choose one […]
Read more "Big-O by a Concrete Example"Parallel algorithms are here! Parallel algorithms are now standard, accessible in VisualStudio 2017 (version 15.8). According to a wonderful Microsoft blog C++17 parallel algorithms are no longer experimental. Algorithms such as sort, for_each, reduce, equal, count, and many more. This give us a standard and portable way to use all of the cores in a multi-core processor. […]
Read more "Standard Parallel Algorithms Have Arrived"
Algorithm Performance 