Re: GETF Speedup?
"Tim Bradshaw (as tfb at tfeb dot org)" <[email protected]>
| Newsgroups | gmane.lisp.lispworks.general |
|---|---|
| Message-ID | <[email protected]> |
On 16 Oct 2025, at 11:04, Yuri Davidovsky (as work at disclosure dot ie) <[email protected]> wrote: > > 1. What is GUPS? Giga Updates Per Second: it's a measure of how rapidly a machine can update *random* memory locations. So it's the worst case for caching. Even small amounts of completely random memory updates can start to dominate performance. > 2. In which ways caching can hurt (especially given the current speed gap between CPU and RAM in modern machines)? If you have a lot of uncacheable memory behaviour then you can end up with caches fetching things that they will never need. The answer is (a) don't have that and (b) if you must have that tell the caches to ignore them. > 3. Why does Common Lisp code benefit less from smart compilers than imperative languages (I assume C is meant here as an imperative language)? There were a bunch of papers in the 1990s explaining why this is not, in fact, true. I can probably find some. > 4. What are the von Neumann bottlenecks that need to be avoided and why alternative architectures are need for that? It's basically the problem of limited bandwidth between a processor and its memory. That's why high-performance machines don't look like that any more. > 5. In which way control over hardware is superior to control over compilers? (I am not sure the phrasing is connect as it feels similar to saying “in what ways water melons are superior to vacuum cleaners” but that is what the quoted text kinda says, it seems.) I'll leave that to someone who thinks it's true. _______________________________________________ Lisp Hug - the mailing list for LispWorks users [email protected] http://www.lispworks.com/support/lisp-hug.html