Alan Jay Smith

EECS Department
University of California, Berkeley
Technical Report No. UCB/CSD-84-219
December 1984

http://www2.eecs.berkeley.edu/Pubs/TechRpts/1984/CSD-84-219.pdf

The directions for computer system development over the next five to ten years are surveyed. The type of system that will come to dominate, or be preferred, over that period will be locally distributed; workstations and intelligent nodes will be connected with local area networks, and will be supported with file servers and larger central computer facilities. Levels of reliability and fault tolerance will increase. Standard software systems, such as Unix, will become more prevalent. The performance of high end computer systems will continue to advance, but at a moderate rate. Breakthroughs in multiprocessor performance are unlikely, but multiprocessing (for small numbers of processors) will become more common due to needs for more computation. Specialized hardware will take a more prominent role in certain areas, such as artificial intelligence and graphics.

Throughout the discussion, the hard problems, the research problems and the engineering problems are indicated. Some industrial opportunities are indicated.

BibTeX citation:

@techreport{Smith:CSD-84-219,
    Author = {Smith, Alan Jay},
    Title = {Trends and Prospects in Computer System Design},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {1984},
    Month = {Dec},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/1984/5928.html},
    Number = {UCB/CSD-84-219},
    Abstract = {The directions for computer system development over the next five to ten years are surveyed. The type of system that will come to dominate, or be preferred, over that period will be locally distributed; workstations and intelligent nodes will be connected with local area networks, and will be supported with file servers and larger central computer facilities. Levels of reliability and fault tolerance will increase. Standard software systems, such as Unix, will become more prevalent. The performance of high end computer systems will continue to advance, but at a moderate rate.  Breakthroughs in multiprocessor performance are unlikely, but multiprocessing (for small numbers of processors) will become more common due to needs for more computation.  Specialized hardware will take a more prominent role in certain areas, such as artificial intelligence and graphics.  <p>  Throughout the discussion, the hard problems, the research problems and the engineering problems are indicated. Some industrial opportunities are indicated.}
}

EndNote citation:

%0 Report
%A Smith, Alan Jay
%T Trends and Prospects in Computer System Design
%I EECS Department, University of California, Berkeley
%D 1984
%@ UCB/CSD-84-219
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/1984/5928.html
%F Smith:CSD-84-219