Read more
Supercomputing refers to exclusive processing at the outer limit of computability - controversial, politically attractive, and very expensive.David Gugerli and Ricky Wichum explore the development of supercomputing in Stuttgart since 1970, and the surprising twists, operational crises, and new technologies it entailed. For example, who would have expected that expansion of Stuttgart's computing center in the 1970s would be capped off with the installation of an outdated supercomputer? Or that the spectacular acquisition of the world's fastest computer in the 1980s would be followed by a years-long quest for users and suitable forms of operation? When, in the 1990s, the Internet made global connectivity possible, Stuttgart was at the forefront, flaunting its dominance in a display of transatlantic experiments. Yet the practical question of what to do with supercomputing was ultimately decided at home, in Germany. The proper management of "users" and extending services to Europe occupied much of the 2000s. By then, previously unanticipated limits to growth had become apparent in the hardware.Told from a history of technology perspective, this study shows that productive supercomputing requires the constant reconfiguring of computers, science, industry, and policy.
List of contents
User's guide The centrality issue (1972-1987) Gaining dominance Planning crisis and a flood of proposals Attempted resuscitation Shaping policy and organizational structure A diversity of machines Shielding users from complexity Communicating to the public The performance gambit (1988-1996) Simulation for all The cost of visualizing computing output The false security of benchmarks Autonomy through regional cooperation Stuttgart's two-pronged solution Network to the rescue (1997-2005) Feasibility study for a national high-performance computing network Metacomputing - a transatlantic experiment Does the university really need an HLRS? Grid computing extends a lifeline Users at work (2006-2016) Taking it easy With Gauss to Europe Virtual users, and users in virtual reality Limits to growth A history of reconfiguration Acknowledgments Notes Bibliography List of figures
About the author
David Gugerli was born in 1961. He is a professor of history of technology at ETH Zurich. His research focuses on the question
of how the world got into the computer.Ricky Wichum was born in 1982. He is teaching the history of technology at ETH Zurich. His research focuses on the historical sociology of computer-based information systems in public administration.
Summary
Supercomputing refers to exclusive processing at the outer limit of computability – controversial, politically attractive, and very expensive.
David Gugerli and Ricky Wichum explore the development of supercomputing in Stuttgart since 1970, and the surprising twists, operational crises, and new technologies it entailed. For example, who would have expected that expansion of Stuttgart’s computing center in the 1970s would be capped off with the installation of an outdated supercomputer? Or that the spectacular acquisition of the world’s fastest computer in the 1980s would be followed by a years-long quest for users and suitable forms of operation? When, in the 1990s, the Internet made global connectivity possible, Stuttgart was at the forefront, flaunting its dominance in a display of transatlantic experiments. Yet the practical question of what to do with supercomputing was ultimately decided at home, in Germany. The proper management of “users” and extending services to Europe occupied much of the 2000s. By then, previously unanticipated limits to growth had become apparent in the hardware.
Told from a history of technology perspective, this study shows that productive supercomputing requires the constant reconfiguring of computers, science, industry, and policy.
