How GitLab decreased repo backup times from 48 hours to 41 minutes with a fix to Git
How GitLab decreased repo backup times from 48 hours to 41 minutes with a fix to Git
Learn how we tracked a performance bottleneck to a 15-year-old Git function and fixed it, leading to enhanced efficiency that supports more robust backup strategies and can reduce risk.
Nah, I was excited to read about the algorithmic change, but it turned out to be an obvious change. I would replace nested loops with a map too. The result is impressive, though.
Marketing departments love to make a huge deal out of this kind of thing, because they only see the big number improvement and don't really understand that this was just some dev's Wednesday afternoon.
And they are right to do so. In the grand scheme of things, it doesn't really matter how much time you spend on a problem. It's the result that matters. I remember a meme where a dev would place a "wait" function in a new feature. Than remove the wait call and call it a free update and get lots of praise from the customer.
I mean, it’s still really impressive upgrade even if technically it was a simple change, they are right to make a fuss about the change
Well it's a 70 times improvement so the developer gets a 70 times bonus. Or at least all the money that would be wasted without this fix.
Or is the world unfair and are developers nowadays just cogs in a capitalistic machine?
Worst bugs usually hide in the most trivial causes 😭
How cool! This is one great point of FOSS.
we traced the issue to a 15-year-old Git function with O(N²) complexity and fixed it with an algorithmic change, reducing backup times exponentially.
I feel like there is something wrong with this sentence.
I'm not a native speaker, but would agree that it sounds imprecise. To my understanding, that's a polynomial reduction of the time (O(n^2) to O(n): quadratic to linear) and not an exponential speed-up (O(2^n) to O(n): exponential to linear). 🤷 Colloquially, "exponentially" seems to be used synonymously to "tremendously" or similar.
Seem ok to me, both in grammar and what it's saying about the change. O(N²) to O(N) would be an exponential drop (2 down to 1, in fact).
An "exponential drop" would be a drop that follow an exponential curve, but this doesn't. What you mean is a "drop in the exponent", which however doesn't sound as nice.
It's at least misleading 😛
But I have to agree that for any non-math people this would convey the right idea, whereas "quadratic improvement" would probably not mean anything 🤷
They make the same mistake further down the article:
However, the implementation of the command suffered from poor scalability related to reference count, creating a performance bottleneck. As repositories accumulated more references, processing time increased exponentially.
This article writer really loves bullet point lists, too. 🤨
There isn't. This is the colloquial use of "exponentially" which is very obvious from the context.
Interesting how it had to get to 48 hours before someone pulled out a profiler