Re: Intel Moves Towards Itanium and Xeon Convergence

n.rieck@xxxxxxxxxxxx wrote:
Intel Moves Towards Itanium and Xeon Convergence

http://www.dailytech.com/article.aspx?newsid=6236

Nothing new here: as the actual interviewer observed (though the article *about* the interview seems to have missed), having a common multi-socket interconnect (which has been a public goal since 2004) *implies* a similar cache architecture (in order to be able to perform cache coherence operations over the same interconnect), plus, of course, to a large degree cache (save perhaps L1) is cache and there's little reason for Intel not to leverage the same tricks in both platforms.

- bill
.

Relevant Pages

  • Re: Pre-query data and load into memory!!??
    ... The CACHE statement is not supported in AS 2005 and should be removed from ... I'm not sure what the interview was talking about. ... typical queries to the cube. ...
    (microsoft.public.sqlserver.olap)
  • Re: Digital Symphony on StrongARM
    ... implies that they did, but I don't remember hearing from them about it ... (I was a registered user and had previously bought an upgrade). ... I have managed to patch it so that it loads with the cache on. ... still won't load packed files with the cache on, ...
    (comp.sys.acorn.apps)
  • Re: Digital Symphony on StrongARM
    ... Their post to comp.sys.acorn.announce in 1997 ("Digital Symphony Now SA ... implies that they did, but I don't remember hearing from them about it (I ... is there a patch so that the cache does not need to be turned off ... when loading the application or compressed tracks? ...
    (comp.sys.acorn.apps)
  • L1 Cache
    ... I am doing research related to L1 cache. ... 1)I would like to know is the block addressing mechanism in L1 ... such that the CPU can send both L1 cache request and TLB ... this implies 128/4=32 ...
    (comp.arch)