Last Friday, Dr Phillip Bording of Nichols Research gave us a very nice seminar on the computing environments (hardware).

At first, Dr Bording introduced the contents of the lecture: 1. Limits of hardware. 2. Limit of software. 3. Application.

Then Dr Bording presented the comparision of CMOS, TTL, ECL with the parameter of progpagation deley in the 20 years before and the application of digital signal processing to radar.

Then Dr Bording presented the model of memory and some algorithms of word counting (WC): unoptimized WC, table driven WC, WC without point, WC with bitwise logistical operation, WC with a pair of reverseing variable (3 time faster), and assemble language WC.

Then Dr Bording introduced the Penta-flop computing: sizing the seismic problem to fit the computing resourse. Seismic imaging is the inverse problem.

Then Dr Bording presented the concept of the wave euation and the seismic modeling. The seismic modeling includes: spectral method, finite element method, finite deference method, etc.

Dr Bording then presented the complexity of seismic modeling and seismic image and the summation: 1. Wave equation imaging is essential for seismic. 2. Smaller exploration targets will require higher resolution. 3. wave machine are "experimental" but feasible solution.

At last, Dr Bording presented some of the computational models: Cache solution, Memory cache/registers inter-connections, Direct memory/registers inter-connections, etc.

Last Update: 3/23/98
Web Author: Zizhong Wang
The report is for Dr Paprzycki@ marcin.paprzycki@ibspan.waw.pl or@ Home Page