3 ms·
Frankly, the problem described is not a problem at all, as it would be weird to rely on cache hits to ensure performance of linked lists. Cache misses are an ex
by perdunov 12y ago
Frankly, the problem described is not a problem at all, as it would be weird to rely on cache hits to ensure performance of linked lists. Cache misses are an expected drawback of linked lists, as they have other advantages.
- ColinWright 12y agoI would think that if you were to ask people about a sudden and unexpected 9-fold drop in performance, they would say that it was a problem. Certainly to me, never knowing when your system might slow down by three orders of magnitude (base 2, nearly one order of magnitude base 10) is a problem.
- perdunov 12y agoYes, I agree that this is a peculiar case and it is worth reading and keeping in mind that such things can happen. I am just saying that I think it would be more correct to consider this as a 9-fold performance increase gained from caching, as no one should rely on caching when dealing with linked lists.
- MichaelGG 12y agoSure, it's a problem because the programmer used a completely unsuitable algorithm that is slow for this use case regardless of memory layout. It's similar to looping over a predicateless SELECT instead of using COUNT. And then blaming databases because I misunderstood the "promises" provided. Although, the original article doesn't make the claims that your submitted title did. (No hostility, just pointing out that you're adding a wrong interpretation.)