Message-passing

Beazley D M and Lomdahl P S 1993 Message-passing multi-cell molecular dynamics on the Connection Machine 5 Parallel Comput. 20 173-95... 

Berendsen, Van der Spoel, D. Van Drunen, R. GROMACS A message-passing parallel molecular dynamics implementation. Comput. Phys. Comm. 91 (1995) 43-56. 

Our multipole code D-PMTA, the Distributed Parallel Multipole Tree Algorithm, is a message passing code which runs both on workstation clusters and on tightly coupled machines such as the Cray T3D/T3E [11]. Figure 3 shows the parallel performance of D-PMTA on a moderately large simulation on the Cray T3E the scalability is not affected by adding the macroscopic option. 

Two implementation decisions could be made immediately. First, DPMTA is based on the PVM message-passing library [11] and therefore it was necessary to base NAMD on PVM as well. All communication done by NAMD, however, would use an intermediate interface to allow communications to be easily retargeted to MPI [12] or other standards, and to simplify later implementation of communication optimizations such as combining messages destined for the same processor. Second, after much debate C++ was selected... 

W. Gropp, E. Lusk and A. Skjellum, Using MPI Portable Parallel Programming with Message-Passing Interface, Scientific and Engineering Computation series, 1994. 

The nodes in a SOM are drawn with connections between them, but these connections serve no real purpose in the operation of the algorithm. In contrast to those in the ANN, no messages pass along these links they are drawn only to make clear the geometry of the network. 

The current protot3rpe system includes three Expert modules, the IR Expert, the STIRS Expert, and the Human. All modules are written in Lisp. The IR Expert is a rule-based infrared interpreter which we have developed. The STIRS Expert is an interface to the STIRS program, a pattern-matching mass spectrum interpreter developed by McLafferty and coworkers at Cornell University, which is written in Fortran. () ) The interface translates the output of STIRS into a form palatable to our program, and handles the message-passing protocol required by the Controller. The Human module controls communication with the user. It allows user-supplied elemental or substructure information to influence the course of the analysis. The power of... 

The nervous system consists of two main units the central nervous system (CNS), which includes the brain and the spinal cord and the peripheral nervous system (PNS), which includes the body s system of nerves that control the muscles (motor function), the senses (the sensory nerves), and which are involved in other critical control functions. The individual units of the nervous system are the nerve cells, called neurons. Nenrons are a nniqne type of cell becanse they have the capacity to transmit electrical messages aronnd the body. Messages pass from one nenron to the next in a strnctnre called a synapse. Electric impnlses moving along a branch of the nenron called the axon reach the synapse (a space between nenrons) and canse the release of certain chemicals called neurotransmitters, one of which, acetylcholine, we described earlier in the chapter. These chemicals migrate to a nnit of the next nenron called the dendrites, where their presence canses the bnild-np of an electrical impnlse in the second nenron. 

A more detciiled account of the parallelization may be of interest to researchers seeking to parallelize electronic structure codes. We consider therefore in this section those aspects of the parallelization that are not directly related to the direct Cl algorithm discussed in Section 2. We first state our overall implementationaJ objectives, followed by some more practical details, including also a description of the PVM [17,18] message passing facility used. 

Figure 2 Speedup for CISD test calculations on CO using the public domain PVM message passing interface.

We show in Figs. 2 and 3 the speed up for a Davidson iteration obtained with the PVM [17] ( Parallel Virtual Machine ) and PVMe [18] ( PVM enhanced ) message passing interfaces respectively. Speed-up factors are here relative to the sequential version of the program (n=l), and the theoretical mciximum has been defined according to the expression s n) — n — I appropriate to the used master/slave model. 

Information and resources relating to the Message Passing Interface (MPI) project is available at http //www.netlib.org/mpi/. ... 

Caughey et al. 1998] S.J. Caughey, M.C. Little and S.K. Shrivastava. Checked Transactions in an Asynchronous Message Passing Environment, in 1st IEEE International Symposium on Object-Oriented Real-time Distributed Computing, pp. 222-229,... 

Due to the nature of FFT, or spectral transforms in general, parallelization of the solver is more suited to a shared memory architecture than to the Message Passing Interface approach. Further, since the most time-consuming portion of the code deals with calculating the RHS of the equation, we choose to apply a single threaded library for ODE solvers (gsl in our case) and only parallelize the calculation of derivatives needed in the ODE solver. 

Only after entering the limbic system does the olfactory message pass to the cerebral hemispheres, which form the major part of the human brain, where cognitive recognition occurs and the ability to associate the smell with its name. By this sequence of events our full awareness of a smell takes place only after the deepest parts of the unconscious have been activated. As a result, even without conscious recognition, smell can be the most evocative of our senses, linking us to past experiences and stirring our emotions at a level that we find hard to explain. 

---