Excessive power consumption emerged as a major obstacle to achieving exascale performance in next-generation supercomputers, creating a need to explore new ways to reduce those requirements. In this study, we present a comprehensive empirical investigation of a power advantage anticipated in the mergesort method based on identifying a feature expected to be physically power efficient. We use a highperformance quicksort as a realistic baseline to compare. Results show a generic mergosort to have a distinct advantage over an optimized quicksort lending support to our expectation. They also help develop some insights toward power efficiency gains likely meaningful in a future exascale context where trading some of the abundant performance for much needed power savings in a ubiquitous computation may prove interesting.

No abstract

Matrix multiplication is ubiquitous in high-performance applications. It will be a significant part of exascale workloads where power is a big concern. This work experimentally studied the power efficiency of three matrix multiplication algorithms: the definition-based, Strassen’s divide-and-conquer, and an optimized divide-and-conquer. The study used reliable on-chip integrated voltage regulators for measuring the power. Interactions with memory, mainly cache misses, were thoroughly investigated. The main result was that the optimized divide-and-conquer algorithm, which is the most time-efficient, was also the most power-efficient, but only for cases that fit in the cache. It consumed drastically less overall energy than the other two methods, regardless of placement in memory. For matrix sizes that caused a spill to the main memory, the definition-based algorithm consumes less power than the divide-and-conquer ones at a high total energy cost. The findings from this study may be of interest when cutting power usage is more vital than running for the shortest possible time or least amount of energy.

No abstract

As exascale systems come online, more ways are needed to keep them within reasonable power budgets. This study aims to help uncover power advantages in algorithms likely ubiquitous in high-performance workloads such as searching. This study explored the power efficiency of binary search and its ternary variant, comparing consumption under different scenarios and workloads. Accurate modern on-chip integrated voltage regulators were used to get reliable power measurements. Results showed the binary version of the algorithm, which runs slower but relies on a barrel-shifter circuit, to be more power efficient in all studied scenarios offering an attractive time-power tradeoff. The cumulative savings were significant and will likely be valuable where the search may be a substantial fraction of workloads, especially massive ones.