Monthly Archives: February 2013

A Picture’s Worth a Thousand Work Items

In the sample code section of this website, you may download code that I have written titled “throughput”. These pictures were produced by this code.

 

The throughput code attempts to measure the performance (or throughput) of a single queue with varying numbers of producer and consumer threads, in search of that “sweet spot” that maximizes throughput through the queue. To make this simulation realistic, producer and consumer threads are given both work and I/O to perform. The amount of work and I/O given to each thread varies according to a normal (Gaussian) distribution. For a description of the code, see: https://heidmann.com/?page_id=15.

The above pictures are from identical runs (that is, identical input parameters), with one run executing on my quad core intel i7, and the other executing on my raspberry pi. The color codes on the right of each picture indicate the number of work items that actually got through the queue. The X and Y axis are the number of consumer and producer threads.

One of the things that is very interesting (and a bit unexpected) to me is that the “sweet spots” are not nearly as far apart as I would have expected (with respect to the number of producer and consumer threads).

Comments, anyone?

Raspberry Pi

This weekend I finally got a raspberry pi, and I have to admit, I’m a bit excited. Since the beginning of my career as a software engineer, most of my time has been spent developing embedded software (some of it real time, some of it safety critical). My current employer, iTRACS, doesn’t do any embedded work, so this little “dev board” (that’s what we call boards purposed specifically for development – playing around, really) will give me the opportunity to keep my embedded skills from getting all rusty.

 

The raspberry pi is a computer, albeit a very small and cheap one (and it, for all practical purposes, is an embedded computer). The whole thing is only a bit larger than a credit card, has a 700MHz ARM processor, an ethernet port, two USB ports, and GPIO (general purpose I/O) pins (and a few other things). It does not have a disk drive, but rather boots of off an SD card (yup, the same ones that cameras use). My raspberry pi is currently running an optimized Debian Linux called “Rasbian”. It was one of the pre-built images available from raspberry pi. It will do for now (but who knows what I’ll do later).

Once I got the board off of its back (that is, got it to boot), I went about getting a cross tool chain working on my desktop Linux box. This will allow me to build (compile) programs on my big, beefy desktop and transfer the resulting executable to my “razz” directly. It is much better build on my desktop than to attempt it on the little pi.

Also, since I am a C++ developer, I make frequent use of the boost libraries. These I also had to cross compile (using my shiny new cross tool chain). It took a bit of doin’, but I eventually figured it out. I went ahead and built some C++ code I had previously developed, and, after downloading the new boost libraries to my “razz”, the code executed without a hitch!

I love it when a plan comes together!