CDRIW@cr83.staffordshire-polytechnic.ac.uk (10/16/89)
> The whole point of memory is that it remembers things over an > indefinite period. I take it you mean RAM in this context. This would be true if there was a realistic, widely available alternative to RAM. Vast amounts of von-neumann RAM is so damn useful. People spend years learning how to map all sorts of inappropriate problems to this architecture. The whole point of RAM is that there is lots of it and people are used to it. This doesn't necessarily make it a good vehicle for pattern matching, or whatever (rather like FORTRAN really). > You use it precisely because you don't want to be > working on all your data all the time - some bits of data (most of it) > lie unused for long periods. So it doesn't look at all obvious that > you should aim to keep all your memory as BUSY as possible. I would say that apart from execution time, it isn't at all obvious what data lies around waiting for. Stuff that is unused probably lives on disk anyway. > I would rather keep it as FULL as possible of useful data, organised so > that individual items can be rapidly located and fetched when they are > needed. If you have some purpose in mind for which BUSYNESS rather than > ACCESSIBILITY ... Are these mutualy exclusive? Indeed, when enough processing is available to swamp the RAM, all sorts of 'look-ahead' algorithms come to mind. > Once the data is available, some thread of the program ought to get > to work to process it straight away. Indeed. > 1: If it could be made to work, there would be no need to use "memory" at all; > when an item of data became available, it could be stuffed straight into > whatever mechanism was going to "process it straight away"; Indeed. (A *very* big If though.) Why must memory be seen as a static machine? Could not the past state be recorded in some way by flowing data? > 2: it would only work on the (totally unrealistic) assumption that ALL > the data required for each stage of the computation would become > available simultaneously. eh? > What IS a real issue is the efficient use of the PATHWAYS to memory - the bus > or whatever; but that's quite a separate question, and fairly well understood. This amounts to much the same thing, wouldn't you agree? Iestyn Walters. (I.D.Walters@stafpol.cr83 if you prefer)