******************************************************************************* Cayenne is a parallel version of Berkeley 2G6 Spice. It utilizes the PlibV2 parallel library routines. PlibV2 is also available through the DECUS Library. Parallel is faster. The purpose of Cayenne is to run a parallel version of Spice 2G6 on any VAX/VMS Multi Processor, which at this time includes the VAX 8300 and the VAX 8800. A set of routines which embeds the parallelization methodology used for CAYENNE and greatly facilitates parallel program development is given in the file CAYEN.FOR. This file with the files PLIBFOR.FOR and PLIBMAC.MAR form the library of routines developped for the CAYENNE methodology. Two SPICE input files are also in the directory: BJTADDER.SPI and MOSADDER.SPI Outputs for these input files are: BJTADDERBST.SPO and MOSADDERBST.SPO These files will verify the CAYENNE application. Benchmarking this application on an VAX 8300 MP, has yielded results from 1.5 to over 1.8 times the single stream version of SPICE. Results will vary due to the size of the data sets. More dense data sets will tend to yield greater performance; hence greater throughput. Several Command procedures have been included for ease of use. All of these procedures share the extension .COM. COMPILE_ALL.COM will compile all the FORTRAN routines that compose CAYENNE. LCN.COM and LCN.COM should be executed, respectively, to gather the and link the object files. CAYENNE is run with 0 slave processes with the command file: RUNIT_SINGLE.COM This will set the number of subprocesses equal to 0 to perform the single stream execution. RUNIT_SINGLE.COM gives the input and output files to CAYENNE. Note: file should be edited for your device and directory specification. The command line that executes CAYENNE in this case is: $ CAY/SUB=0 where CAY is defined by: $ CAY :== $[]CAY.EXE. To run a parallel version of CAYENNE, the file RUNIT.COM can be executed. This file is similar to RUNIT_SINGLE.COM and executes CAYENNE using two subprocesses. The startup command line is $ CAY/SUB=2 Note: file should be edited for your device and directory specification. To avoid timing data at the end of the output file, which will speed up the wall clock time for completion to some extent, rename task_notiming.for to task.for. ************************************************************************ NOTE: CAYENNE may require some adjustments to parameters such as MAXJOBS, GBLPAGFIL etc. to accomodate it's needs. ************************************************************************ ************************************************************************ ************************************************************************