ABSTRACT
The concepts and methods are reviewed which have proven to be of the most value in designing and implementing the Octopus computer network, which is one of the largest concentrations of computing capability in the world. The discussion summarizes design principles relating to the scope of system software, privacy and security, processor organization, file structure, network connections, hardware selection, programming techniques, standards, and use of resources. Differences with what appear to be widespread belief and practice are cited, including such matters as the size of the programmer staff, the significance of the privacy issue, the importance of the choice of languages and language constructions, the primary causes of system failure, and efficiency.
References
- 1.J. G. Fletcher, The Octopus computer network, Datamation, vol. 19, no. 4, April 1973, 58-63.Google Scholar
- 2.J. E. Requa, In-house vs. vendor-supplied software: a case study at Lawrence Livermore Laboratory, Atomic Energy Systems, Operations and Programming Association Meeting, October 1972.Google Scholar
- 3.J. D. Kuehler and H. R. Kerby, A photodigital mass storage system, Proc. AFIPS FJCC, vol. 29, November 1966, 733-742.Google Scholar
- 4.R. C. Daley and P. G. Neumann, A general purpose filing system for secondary storage, Proc. AFIPS FJCC, vol. 27, November 1965, 213-229.Google Scholar
- 5.G. B. Houston, Trillion Bit Memories, Datamation, vol. 19, no. 10, October 1973, 52-58.Google Scholar
- 6.S. F. Mendicino, R. A. Hughes, J. T. Martin, F. H. Mc Mahon, J. E. Ranelletti, and R. G. Zwakenberg, The LRLtran Compiler, Comm. A.C.M., vol. 11, no. 11, November 1968, 747-755. Google Scholar
Digital Library
Index Terms
Principles of design in the octopus computer network




Comments