Editing ETA Systems (section)

==Products==
ETA had only one product, the [[ETA10|ETA-10]]. It was a derivative of the [[CDC Cyber 205]] supercomputer, and deliberately kept compatibility with it. Like the Cyber 205, the ETA-10 did not use [[vector processor|vector registers]] as in the [[Cray Research|Cray]] machines, but instead used pipelined memory operations to a high-bandwidth [[main memory]]. The basic layout was a shared-memory multiprocessor with up to 8 [[Central processing unit|CPUs]], each capable of 4 double-precision or 8 single-precision operations per clock cycle, and up to 18 [[I/O processor]]s.

The main reason for the ETA-10's speed was the use of [[liquid nitrogen]] (LN<sub>2</sub>) cooling in some models to cool the CPUs. Even though it was based on then-current [[CMOS]] technologies, the low temperature allowed the CPUs to operate with a ~7&nbsp;ns cycle time, so a fully loaded ETA-10 was capable of about 9.1 GFLOPS. The design goal had been 10&nbsp;GFLOPS, so the design was technically a failure. Two LN<sub>2</sub>-cooled models were designated '''ETA-10E''' and '''ETA-10G'''.  Two slower, lower-cost air-cooled versions, the '''ETA-10Q''' and '''ETA-10P''' (code named "Piper") were also marketed.<ref>Pricing for the air-cooled systems began at a then unheard of $1,000,000. In fact, they could be purchased and operated at a lower cost than the cost to operate and maintain a low end Cray 1 or Cray-XMP.

What set apart ETA Systems supercomputers and the Cray system was their ability to expand primary memory far beyond that of the Cray machines. Cray had a competitive performance advantage in the scalar processing arena whereas ETA, using large lower cost memory and pipelining hardware design, simply decimated the Cray products where large data sets were in use; requirements such as oil reservoir data analysis and aerodynamic simulations. Operating software at ETA was a major stumbling block to the overall product success. Purdue University successfully installed a copy of UNIX on their ETA-10P. Unfortunately, it was all too late as ETA's parent company, CDC, was rapidly disintegrating and without the funds to continue, ETA Systems died on the vine. Richard Rychtarik Author - Piper Strategic Plan - ETA Systems 3/3/1987.</ref>

The planned successor to the ETA-10 was the 30&nbsp;GFLOPS ETA-30.<ref>{{cite book |last1=Hockney |first1=R.W. |last2=Josshope |first2=C.R. |title=Parallel Computers: Architecture, Programming and Algorithms |date=1988 |publisher=Adam Hilger |page=185 |edition=2}}</ref>