liu@buster.cps.msu.edu (Yew Liu) (01/19/91)
Hi, all ! Could anyone tell me any info about CrOS? I think it was called Crystalline OS. I need to find any article that explains what it is. Thanks in advance Yew
cherwig@hubcap.clemson.edu (christoph herwig) (01/21/91)
In article <12681@hubcap.clemson.edu> liu@buster.cps.msu.edu (Yew Liu) writes: > >Could anyone tell me any info about CrOS? I think it >was called Crystalline OS. I need to find any article >that explains what it is. > You are right: CrOS stands for crystalline OS. CrOS III was designed at Caltech to run efficiently on a specific concurrent processor, the Mark III hypercube. Two books can provide you with startup information about CrOS III: 1. Solving problems on concurrent processors, General Techniques and regular problems; Volume I; G.Fox et al.; Prentice Hall; 1988 2. Solving problems on concurrent processors, Software for concurrent processors; Volume II; I.Angus et al.; Prentice Hall; 1990 Although CrOS III was motivated by particular design of hypercube, it has been written to be easily implemented on other machines. As described in Volume II, it runs particularly efficient on the Caltech/JPL Mark III. For other commercial hypercubes, such as the nCube and iPSc/1 (don't know about iPSc/2 or iPSc/860), it can be implemented simply, but inefficiently, in terms of the native message-passing primitives. On machines with other architectures, such as shared memory multiprocessors, it is implemented by simulating the hypercube communication channels in software. Especially Volume II will be useful and more up-to-data. You will find the description of the implementation on the various machines. The appendices of both volumes give you source code and manual pages of the 'utility library'. Christoph ----------------------------------------------------------------------- Christoph Herwig cherwig@hubcap.clemson.edu Department of Electrical and Computer Engineering Clemson University, SC Elektrotechnik Technical University Aachen, Germany ----------------------------------------------------------------------- technical university Aachen, germany
grunwald@foobar.colorado.edu (Dirk Grunwald) (01/23/91)
re: CrOS et al In particular, one might note that CrOS et al follow the mindset of the phsyics researchers at CalTech, that is, that all communication should be either strictly local (near forces) or global (far forces). Thus, early versions of CrOS only support communication with adjacent neighbours and global broadcast. This was important on e.g., the Mark-III, which was had a store and forward network, but less important on e.g., the Intel iPSC/2 hypercube, where locality was less important. This is not to belittle the CrOS; it was considerably more efficient for localized communication than e.g., Mercury, another Mark-III O/S. (Old Ametek S-14 users will note that Mercury==mars and CrOS==whatever it was that you actually used on the S-14). However, for other than those special purposes, you had to ``roll your own'' multi-hop communication primitives.