Things To Do With a Supercomputer

For the purpose of creating a task list, the definition of supercomputer can be loosely expanded to include clusters, grids, clouds and all forms of conglomeration of a large amount of computing power.

MODELING

  • Weather forecasting – a complex three dimensional fluid dynamics model being constantly recalculated with a massive amount of dynamic data being fed in from sensors
  • Nuclear simulations – a complex three dimensional fluid dynamics model similar to a weather storm simulation, but with larger energy levels, much smaller frameworks of time and space involved that require more recalculations, plus state changes and electromagnetic forces at unusual levels
  • Aerodynamics – a complex three dimensional fluid dynamics model of airflow, first used for aircraft design, but now being used for land vehicles from Formula 1 racing teams to long haul trucks.
  • Molecular dynamics – complex modeling of the physical movements of molecules and atoms requires combinations of many force equations at different levels. This can also be used in bio-genetics and cosmology.
  • Simulation and training – training simulators use a form of virtual reality, but may require more dynamic updating and two way data exchange.
  • Virtual reality – situational models covering a wide spectrum of application from simple to complex and one way to multiple channel directions.
  • Cosmology – complex modeling involving extreme densities, temperatures, gravitation, and bending of both space and time. May also involve molecular dynamics as a subset.
  • Geophysics – modeling that involves: seismology, geo-magnetics, atmospherics, geo-thermal dynamics, oceanography, tectonics, glaciology, petro-physics and more.
  • Bio-genetics – modeling and analysis of complex problems involving genetics, nucleic acids and proteins, molecular biology, virology, bio-chemistry and more.

COMPUTING

  • Cryptography – cryptanalysis, crypto-linguistics, brute force attacks, cypher design, prime numbers, large number factoring.
  • Number theory – prime numbers, zeta function, progressions, group theory, complex analysis, algebraic integers.
  • Large data sets – big data can be generated by many of the applications listed here, but often they are focused on calculations on the data. Data mining, filtering, sorting, querying, arranging and reporting large data sets also requires big resources.
  • Optimization algorithms – simplex algorithm, network optimization, quadratic programming, linear and linear fractional programming.
  • Experimental mathematics – prime number search, puzzle solving, searching for patterns, proof by exhaustion, evaluation of infinites.
  • Evolutionary algorithms – search heuristics, genetic algorithms, evolutionary strategy and programming, learning classification, swarm algorithms, harmony and combination searching.
  • Machine learning – neural networks, decision trees, automata, regression analysis, nearest neighbor, Markov models, Bayesian networks, clustering, self-organizing.

SEE ALSO:
Moores Law
Future Grid
Cubic Centimeter Supercomputers
High Performance Cluster on Amazon EC2
Data Mining
timeline of fluid mechanics

Comments are closed.