eugene@pioneer.arc.nasa.gov (Eugene N. Miya) (06/18/88)
Returning yesterday from the Intl. COnf. on Dist. Comp. Systems, I ran into Computer Literacy Bookstore [San Jose location] to be flooded with a slew of books of parallelism. I broke down the other day and puchased Robbie Babb's book because the Government library service is too slow. It's very survey, a collection of Grad student papers put together, so its shallow, but does give some sense of syntax. The two chapters to watch out for: the IBM 3090 chapter on EPEX is disappointing (alan Karp did all the actual programming, not the students) [How important is it to have "hands on?" I think it's essential after using two of the Japanese machines] [BTW: friends at AT&T and E&S made me go those other days. The Devil made me do it...] The other significant book is Geoffery Fox's vol. 1 of his new book on concurrent processors. This book LOOKS okay, but knowing of his work from the SMP days, I am certain there are some idiosyncracies with this book. There are at least a dozen lesser works with authors I'm unfamiliar with. Most significant perhaps is Mary Mace's book (PhD thesis) on Memory Access Patterns for Parallel Processors. The thing to note about this book is that she works for IBM. I have all these books on order with our library, I would never and could never buy all of them out of my pocket. I also fear that I will miss some of these books since I don't have a pipeline from all these publishers. Another gross generalization from --eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov resident cynic at the Rock of Ages Home for Retired Hackers: "Mailers?! HA!", "If my mail does not reach you, please accept my apology." {uunet,hplabs,ncar,decwrl,allegra,tektronix}!ames!aurora!eugene "Send mail, avoid follow-ups. If enough, I'll summarize."