Gene Amdahl, RIP
Katie Hafnernov (via Slashdot, comments):
Dr. Amdahl rose from South Dakota farm country, where he attended a one-room school without electricity, to become the epitome of a generation of computer pioneers who combined intellectual brilliance, managerial skill and entrepreneurial vigor to fuel the early growth of the industry.
As a young computer scientist at International Business Machines Corporation in the early 1960s, he played a crucial role in the development of the System/360 series, the most successful line of mainframe computers in IBM’s history. Its architecture influenced computer design for years to come.
Computer History Museum (via Grady Booch):
In 1970, Amdahl left IBM for the second and final time to pursue his dream of building his own computers, founding Amdahl Corporation. His new company made mainframe computers that ran IBM software, but at lower cost. At its peak, it captured nearly one-fifth of the market.
FUD was first defined with its specific current meaning by Gene Amdahl the same year, 1975, after he left IBM to found his own company, Amdahl Corp.: “FUD is the fear, uncertainty, and doubt that IBM sales people instill in the minds of potential customers who might be considering Amdahl products.”
Gene Amdahl was the John DeLorean of mainframes (but without the cocaine). Invented, tried to out-compete, the IBM 360.
In computer architecture, Amdahl’s law (or Amdahl’s argument) gives the theoretical speedup in latency of the execution of a task at fixed workload that can be expected of a system whose resources are improved.
[…]
Amdahl’s law is often used in parallel computing to predict the theoretical speedup when using multiple processors. For example, if a program needs 20 hours using a single processor core, and a particular part of the program which takes one hour to execute cannot be parallelized, while the remaining 19 hours (p = 95%) of execution time can be parallelized, then regardless of how many processors are devoted to a parallelized execution of this program, the minimum execution time cannot be less than that critical one hour.
Update (2015-11-14): btilly:
The technical staff said that the operating system should run on microcode to abstract away the hardware. That way it would be easier for customers to migrate to new hardware as it became available. And they could easily add a new instruction if they needed to.
Gene said that it would be an order of magnitude faster if it ran directly on the hardware, and it wasn’t that hard to support that API going forward.
Both proved right. Gene built computers that were massively faster than IBM’s and perfectly compatible. IBM then added an instruction in micro-code and made all of their software use it. Gene’s installed base all crashed on IBM’s new code, while IBM’s was fine. The US government launched an anti-trust lawsuit, which wound up binding IBM’s hands for many years after.
IBM mainframes today still run on micro-code. And it still makes them massively slower than they need to be, but with better backwards compatibility. The mainframe world depends on a lot of programs from the 1960s and 1970s that runs, unchanged, today. Everyone else is using native instructions and runs faster.
“Because as soon as the IBM sales rep sees the Amdahl coffee cup on your desk, he’ll know I was here and he’ll drop his price by $1 million if you ask him to.”
ACM:
Amdahl left his company in 1979 to set up Trilogy Systems, an organization aimed at designing an integrated chip for even cheaper mainframes. When the chip development failed within months of the company’s $60-million public offering, Trilogy focused on developing its VLSI technology, which also did not do well. In 1985 Trilogy was merged into microcomputer manufacturer Elxsi (now Tata Elxsi), but poor results there had Amdahl leaving in 1989 for a company he founded in 1987 to produce mid-sized mainframes, Andor International, which had been driven into bankruptcy by production problems and strong competition by 1995.
[…]
Said David Patterson, a professor of computer sciences at the University of California, Berkeley, and a computer pioneer in his own right, “The IBM System/360 was one of the greatest computer architectures of all time, being both a tremendous technical success and business success. It invented a computer family, which we would call binary compatibility today. When he left to form his own company, his mainframes were binary compatible with the System/360.”
[…]
In addition to Amdahl’s Law, Patterson said, "Less well-known are Amdahl’s rules of thumb for a balanced computer system," which include, "A system needs a bit of IO per second and one byte of main memory for each instruction per second."
