Following a support conversation with a customer on a slightly different issue, we realised that there was potential for a signification optimisation in UpdraftPlus’s zip creation engine, when the default engine is being used.
We’ve now implemented and tested it, and the results are really impressive. Using a 180Mb media library that we had to hand for testing, we found that we could reduce the amount of time spent in zip processing by around 50%.
This kind of optimisation brings wins all around. Less CPU is used, more can be done in less time (and so less resumptions are needed on hosting setups where a process gets killed after a defined number of seconds), backups finish quicker, more can be squeezed out of bargain-basement hosting, etc.; everyone’s happy.
This change is in our development trunk, has been tested, and next time there’s a release (hopefully later this month) – if your server supported the default zip engine (most do), then expect quicker and less resource-intensive backups. And if your site is a really large one, with a big media library, then the bigger it is, the more you’ll be gaining from this improvement!
Oh, and – this is one more pay-off from our policy of having our support desk manned by software engineers. People trained and capable for doing support only can keep costs down and maximise today’s profits; but engineers can spot patterns, understand deeper issues, and work out how to turn a simple support question into a gain for everyone. We’ve followed this policy since day one (when it was just me, before we had others), and have no plans to stop.
David Anderson (founder, lead developer, UpdraftPlus)