3 ms·
His response just loaded on my end for some reason. Yeah, that's the points I was getting from his longer write-up. If you see him redo it & I'm not there, defi
by nickpsecurity 10y ago
His response just loaded on my end for some reason. Yeah, that's the points I was getting from his longer write-up. If you see him redo it & I'm not there, definitely ping me so I can see the future copy.
We were actually both in the best example I've seen of this effect on HN:
https://news.ycombinator.com/item?id=12472905 https://news.ycombinator.com/item?id=12472905
It's the kind of title that, if proven in resulting article, should make many people pause to think about how effectively they're solving business problems.
- qwertyuiop924 10y agoAh, I remember that. To paraphrase myself at the time, "never send a 747 to do a racecar's job." Yeah, as it turns out, it's not a good idea to use a piece of software optimized for workloads to large to even be handled by a single machine on large quantities of data, and maybe you should use tools designed for small workloads (Gigabytes, or even Terabytes, is pretty small, nowadays) for small workloads. This should be obvious. Besides, things like Manta exist, so you can scale up your small workload scripts to larger workloads in the future. Mind, I don't know if I trust Manta yet (I don't know the perf, and I haven't tried it personally), so maybe exercise a little bit of caution about going that route.