jack@cwi.nl (Jack Jansen) (02/26/88)
I'm looking for articles describing algorithms for deciding when to move data from one cache to another in a multiprocessor system. It seems that mosts systems use something along the lines of "as soon as you access the data you get a copy", but I was wondering whether there are smarter ways to decide in which cache to store the item, especially in cases of shared variables with more than one processor having write access to it. Could anyone give me a lead to papers that could help me along? -- Jack Jansen, jack@cwi.nl (or jack@mcvax.uucp) The shell is my oyster.