Fast Gaussian-distributed Random Numbers
Recently, profiling showed me that most of the time in the microbenchmark I was running was spent generating the stream of random numbers using Random.nextGaussian. »
So, Patrick, how was your long weekend?
I took this past Monday as a vacation day. Upon returning, I sent this email to the team, and they suggested I post it here. Upon »
Help! I invented an algorithm and I don't know how it works
This past weekend, I needed to find a good fraction to approximate a given value. Along the way, I learned about continued fractions, and emerged with »
The Case for a Logarithmic Performance Metric
Traditional comparisons of benchmark results focus on percentage differences. While seemingly simple to understand and compute, such comparisons pose a number of problems that are solved »
Measuring information in millibytes
A thousandth of one byte. Does such a thing even exist? Yep, it does. I've used them before. How to debug an optimizing compiler Suppose you »
Gotcha! Cartesian products of Java 8 streams
Occasionally, you need to compute the Cartesian product of several sets. Java 8 Streams make this simple, clear, elegant... and extravagantly expensive. I've created a Cartesian »
OCaml mimicry in Java
Yaron Minsky from Jane Street gave a great talk on OCaml that serves as a good introduction for someone who is accustomed to Java. In particular, »
The Inflatable Transpose Tree: a highly compact data structure
In honour of Claude Shannon's 100th birthday, I want to tell you about two ideas I had for storing data efficiently. These ideas are pretty nifty »
Malloc envy
Folks using programming languages with automatic garbage collection (GC) often feel that it's a price worth paying. I disagree: outside of some fairly limited circumstances, there's »
Technical Talk at University of Toronto on March 23
A few of us will be visiting U of T next week to give a talk on Vena's Calc programming language. If you like programming languages »