Hacker News Reader

Comment by warent | Hacker News Reader

No offense intended at all but I can't help but wonder how a technologist/computer scientist has ended up asking for help resolving 500 errors on his blog rather than hypothesizing what the problem could be and testing his theories.

A quick Google search on the error he's getting says it's an Apache error either from invalid file permissions or running out of memory. His provider refuses to transparently increase memory so the hypothesis seems stronger here. Now just test it by spinning up a simple load balancer, clone your website on a different server, split the traffic 50/50, and see if the errors continue. Repeat this process until you've isolated the issue. Trying to solve the problem by adding a cache layer isn't going to help before the problem is understood.

View on HN

We don’t know that the author didn’t google and try a few things. And it’s perfectly possible to be a computer scientist and not know how to tune a web server. There are lots of specialisms and paths. Personally I admire someone who, despite being an expert, isn’t afraid to ask questions and admit when they don’t know something.

That's a great point. My comment above seems to be coming from a rather limited mindset

It’s getting more funny if you look at his GitHub account.

https://github.com/lemire “His research is focused on software performance and indexing.”

I think it’s just a poor advertising. :)

Looking at one of the papers he has published

"Many applications use sequences of n consecutive symbols (n-grams). Hashing these n-grams can be a performance bottleneck. For more speed, recursive hash families compute hash values by updating previous values. We prove that recursive hash families cannot be more than pairwise independent. While hashing by irreducible polynomials is pairwise independent, our implementations either run in time O(n) or use an exponential amount of memory. As a more scalable alternative, we make hashing by cyclic polynomials pairwise independent by ignoring n-1 bits. Experimentally, we show that hashing by cyclic polynomials is is twice as fast as hashing by irreducible polynomials. We also show that randomized Karp-Rabin hash families are not pairwise independent."

I think he is working on the fundamental science of computer science. People like him are the reason we have nice things like web servers. That doesn't mean they know the specifics of Apache or Wordpres.

Reply on news.ycombinator.com