Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.programming.threads > #6190

Re: Read again, i correct a typo about hash function collision and scalability..

Show all headers | View raw

> in my following software projects can attain only a probability of 50% for a collision at around 77000 cores and and 77000 threads, so it is good for scalability even if i am not using a good 64 bit hash function,
> (but i will soon upgrade them to a good 64 bit hash function),

You have megalomania. There will never be a computer with 77.000 cores
in shared memory. NUMA won't scale to that extend, M(O)ESI won't scale
to that extent and the software also won't scale to that extent. sgi
has dropped is massively parallel shared memory system with its NUMA-
link-architecture because of that.

Back to comp.programming.threads | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Read again, i correct a typo about hash function collision and scalability.. Amine Moulay Ramdane <aminer68@gmail.com> - 2020-12-22 13:40 -0800
  Re: Read again, i correct a typo about hash function collision and scalability.. Bonita Montero <Bonita.Montero@gmail.com> - 2020-12-22 22:50 +0100

csiph-web