How much Algebra does C2 Know? Part 1: Associativity

Making loops execute faster is firmly rooted in algebra, but how much does C2 know or care about? When building a highly optimised query engine, a critical concern is the quality of assembly code generated for loops. There is a lot more to JIT compilation than loop optimisation; inlining, class hierarchy analysis, escape analysis to […]

Zeroing Negative Values in Arrays Efficiently

Replacing negatives with zeroes in large arrays of values is a primitive function of several complex financial risk measures, including potential future exposure (PFE) and the liquidity coverage ratio (LCR). While this is not an interesting operation by any stretch of the imagination, it is useful and there is significant benefit in its performance. This […]

Still True in Java 9: Handwritten Hash Codes are Faster

One of the things to keep in mind with Java is that the best performance advice for one version may not apply to the next. The JVM has an optimiser which works by detecting intent in byte-code; it does a better job when the programmer is skilled enough to make that intent clear. There have […]

New Methods in Java 9: Math.fma and Arrays.mismatch

There are two noteworthy new methods in Java 9: Arrays.mismatch and Math.fma. Arrays.mismatch This method takes two primitive arrays, and returns the index of the first differing values. This effectively computes the longest common prefix of the two arrays. This is really quite useful, mostly for text processing but also for Bioinformatics (protein sequencing and […]

Choosing the Right Radix: Measurement or Mathematics?

I recently wrote a post about radix sorting, and found that for large arrays of unsigned integers a handwritten implementation beats Arrays.sort. However, I paid no attention to the choice of radix and used a default of eight bits. It turns out this was a lucky choice: modifying my benchmark to parametrise the radix, I […]