|
|
|
|
Mechanical and Aerospace Engineering Applications - 4th Quarter Status ReportWe have only begun to gain experience with the Intel machines and their operating systems in mechanical and aerospace engineering. We proposed work in two general areas: (1) Simulation of dynamical systems and control, and (2) computational modeling of fluid systems. For the former, we plan primarily to use higher-level packages, such as MATLAB, while for the latter we will attempt to develop a networked system of Intel processor machines to perform large calculations in a distributed memory environment. We have received 8 machines to date: 4 have been distributed to faculty working in area (1) and 4 to faculty working in area (2). Some simulation software has been loaded on 2 of the 4 machines for the dynamical systems and control projects, but since the machines only recently arrived (and 2 are just in the process of being set up), there are no results yet to present. We are in the process of setting up 2 of the fluids machines running NT and two running Linux, in order to compare the suitability of the two operating systems for networked parallel computing. Since the network switches arrived only this week, we have no results to present in this area either. Nonlinear Dynamics and Control of Multibody Systems Two workstations are currently being used in two experiments involving real-time control. The first experiment consists of the real time control of an underactuated vehicle (a disk on a low friction surface with two fixed thrusters). The experiment is being used to test new algorithms for controlling vehicles engaged in high-performance and aggressive maneuvers. An overhead camera and associated hardware is used to determine the state of the system. This information is relayed to one of the Intel Workstations which handles the trajectory generation and the construction of the feedback control laws required to follow the trajectory, in real-time. Commands are relayed to the vehicle via a wireless transmitter. The second experiment consists of a three degree of freedom manipulator which is interfaced to an Intel Workstation for real-time control. The purpose of this experiment is to test new algorithms being developed for aggressively controlling manipulators. The other two workstations are being used in the Cornell Robot Soccer project. Two teams of autonomous robots are being constructed by 30 Cornell students. These two teams will engage in a simplified game of soccer. The workstations are being used to process the real-time information being provided by a vision system and to use this information for controlling and coordinating the robots in order to score goals in the opponent's area and defend goals in its own area, in real-time. Details of this competition may be found at http://www.mae.cornell.edu/Robocup/ Computer Modeling of Turbulent Reactive Flows At Cornell, methods for modelling turbulent reactive flows have been pioneered, which are now in use in industry (e.g., in the gas-turbine industry). These methods involve a particle/mesh Monte Carlo method which is computer intensive. Within the reactive flow field being considered, the fluid is represented by a large number N (e.g., N=10^6) particles. These particles have properties that evolve in time according to stochastic differential equations. Typically, of order 1,000 time steps are required. The codes are run on workstations, and in parallel on the IBM SP2. There is a statistical error that scales as N^{-1/2}, and a bias that scales as N^{-1}. It is highly desirable to have access to extensive computer resources so that these errors can be reduced to acceptable levels by increasing the number of particles, N. In industry, the typical computing environment for this kind of calculation is a cluster of workstations. At Pratt & Whitney, routinely 2,000 workstations are used for a single calculation. As PC's become more powerful, there is interest in using clusters of PC's instead. We have been making good use of the Intel PC's to run our particle code. The migration from the SP2 and Unix workstations for serial (single-processor) runs was very straightforward. For a typical serial run, the Intel machines are very slightly faster than a single SP2 node. Recently, 6 of the Intel machines have been combined to form a Beowulf/linux/LAM/MPI cluster. The following table summarizes the performance using 1, 2 and 4 nodes for a preliminary series of runs. For these runs the work per node is kept essentially constant, i.e., the total size of the computation increases linearly with the number of nodes, so the (relative) wall clock time would remain at unity if perfect speed-up were achieved. Number of processors 12 4 Wall clock time (relative) 1 1.65 2.35 Parallel efficiency 160% 42% Speed up 11.2 1.7 Ideal speed up 1 2 4
Mechanical and Aerospace Engineering Applications Status ReportWe have only begun to gain experience with the Intel machines and their operating systems in mechanical and aerospace engineering. We proposed work in two general areas: (1) Simulation of dynamical systems and control, and (2) computational modeling of fluid systems. For the former, we plan primarily to use higher-level packages, such as MATLAB, while for the latter we will attempt to develop a networked system of Intel processor machines to perform large calculations in a distributed memory environment. We have received 8 machines to date: 4 have been distributed to faculty working in area (1) and 4 to faculty working in area (2). Some simulation software has been loaded on 2 of the 4 machines for the dynamical systems and control projects, but since the machines only recently arrived (and 2 are just in the process of being set up), there are no results yet to present. We are in the process of setting up 2 of the fluids machines running NT and two running Linux, in order to compare the suitability of the two operating systems for networked parallel computing. Since the network switches arrived only this week, we have no results to present in this area either.
|
|
Last modified on: 10/08/99 |
