An Efficient Implementation for the BROADCAST Instruction of BSR+
IEEE Transactions on Parallel and Distributed Systems
Towards efficient BSP implementations of BSR programs for some computational geometry problems
EURO-PDP'00 Proceedings of the 8th Euromicro conference on Parallel and distributed processing
Hi-index | 0.00 |
In this paper, we investigate the BSR (Broadcast withSelective Reduction) model of computation for problems related with subsegments (basically, inputs arearrays containing elements and outputs are "somecontiguous part" of inputs), namely MSSP (MaximalSum Subsegment Problem - one and two dimensionalversion) and LIS (Longest Increasing Sequence). TheBSR model of computation introduced by Akl [2] ismore powerful than any CRCW PRAM and yet requires no more resources for implementation than anEREW PRAM. In our solution, we require to add anew reduction operator in the basic brick of the inter-connection unit. The implementation of this operatoris straitforward and does not require more resourcethan those in the original BSR implementation; nofundamental architectural piece of the basic model ischanged.