Abstract: When parallel processing is executed by parallel computers composed of a host computer and a plurality of processors connected to the host computer through a common bus, there is provided a method of assigning jobs to respective processors with high efficiency. A job in which a ratio between a communication time and a calculation time is larger than a predetermined value or larger than a fraction of processors and a job in which a ratio between a communication time and a calculation time is smaller than a predetermined value or smaller than a fraction of processors can be alternately assigned to respective processors.

Abstract: Matrix element calculation carried out efficiently without the overhead of communication between a host computer and processor elements even in parallel calculation utilizing a low-cost communication device and multiple processor elements having memories of a small capacity. In a method for calculating molecular orbitals, for example, all elements F(I, J) of a Fock matrix are calculated where an outermost loop is a loop associated with combinations (RT) of contracted shell R and contracted shell T which satisfy relationships R≦Nshell and T≦R. A second loop is a loop associated with contracted shell S, and a third loop is a loop associated with contracted shell U. Alternatively, the second loop is a loop associated with the contracted shell U, and the third loop is a loop associated with the contracted shell S. The value of S ranges from 1 to R, and the value of U ranges from 1 to R.

Abstract: When parallel processing is executed by parallel computers composed of a host computer and a plurality of processors connected to the host computer through a common bus, there is provided a method of assigning jobs to respective processors with high efficiency. A job in which a ratio between a communication time and a calculation time is larger than a predetermined value or larger than a fraction of processors and a job in which a ratio between a communication time and a calculation time is smaller than a predetermined value or smaller than a fraction of processors can be alternately assigned to respective processors.

Abstract: There is provided a parallel computer and a parallel computing method which allows high precision parallel calculation to be executed without requiring a hardware scale while maintaining high calculation speed. A system is constructed by connecting a host processor with a plurality of special purpose processors via buses. The host processor carries out the operation in a format of double-precision floating-point and the special purpose processor carries out the operation in an internal format of floating-point. The special purpose processor comprises an input data converting section for converting from the double-precision to the internal format and an output data converting section for converting from the internal format to the double-precision. Because the sign part and the exponent part can use data in common in the data before and after the conversion, only the mantissa part is converted by a specific procedure.

Abstract: This invention concerns a belt conveyor for delivering pantyhose whose panty part only has just been seamed by a panty part sewing machine to a toe part sewing machine in order for the leg part to be received smoothly. When the half made pantyhose are on their way to the toe part sewing machine with the panty part held by clipping members and the leg parts dangling from a guide pipe of the panty part sewing machine, the belt conveyor, disposed between both the paths of clipping members of the panty part sewing machine and suction members of the toe part sewing machine, correctly receives the leg parts one by one in association with a guide member disposed in contact with the belt conveyor at a little time interval. One of the leg parts, received by the belt conveyor first, is delivered, by their suction, to the suction members then coming nearby; the other is delivered in the same way a little later.