Sequence search on a supercomputer

Carbon-based nanotechnology on a supercomputer

The quantum nature of phenomena dominating the behaviour of nanostructures raises new challenges when trying to predict and understand the physical behaviour of these systems. Addressing this challenge is imperative in view of the continuous reduction of device sizes, which is rapidly approaching the atomic level. Since even the most advanced experimental observations are subject to being funda...

Parallel Clustering on a Commodity Supercomputer

Fitting the universe on a supercomputer

tens of billions to form vast galaxies, each with a complex, but essentially regular, internal structure extending over tens of thousands of light-years. The distances between galaxies are typically a hundred times greater, yet galaxy-distribution maps show—even on such huge scales—dramatic and obvious structures. Many galaxies are in clusters, the most populous containing a thousand or more sy...

A Streaming Supercomputer

We are in an era where computational building blocks are plentiful and inexpensive. A single chip today can hold over 100 1GHz °oating-point units for a total performance of 100 GFLOPS/chip. Many graphics chips achieve 80GFLOPS and over 1TOP rendering performance, and cost less than $100. Embedded processors are less powerful, but incredibly cheap. It is fair to say that a raw GFLOPS costs less...

A Compiler-driven Supercomputer

The overall prrformance of supercomputers is slow compared to the speed of their underlying logic technology. This discrepancy is due to several bottlenecks: memories are slower than the CPU, conditional jumps limit the usefulness of pipelining and pre-fetching mechanisms, and functional-unit parallelism is limited by the speed of hardware scheduling. This paper describes a supercomputer archit...