TY - CONF T1 - Evaluating techniques for exploiting instruction slack T2 - Computer Design: VLSI in Computers and Processors, 2004. ICCD 2004. Proceedings. IEEE International Conference on Y1 - 2004 A1 - Yau Chin A1 - John Sheu A1 - David Brooks AB - In many workloads, 25% to 50% of instructions have slack allowing them to be delayed without impacting performance. To exploit this slack, processors may implement more power-efficient, longer latency pipelines or provide dynamically scaled pipelines using multiple clock domains. Issuing instructions with slack to slower pipelines can result in substantial power savings, with minimal performance loss. Considering both dynamic and static power dissipation, we found that by using longer latency pipelines the power of functional unit pipelines decreases by 20% to 55% with a performance impact of 0% to 3% for SPEC2000 and MediaBench workloads. Dynamic scaling reduces the performance loss in intense multimedia workloads by up to 2%, but achieves lower power savings. JF - Computer Design: VLSI in Computers and Processors, 2004. ICCD 2004. Proceedings. IEEE International Conference on PB - IEEE CY - San Jose, CA, USA UR - 10.1109/ICCD.2004.1347949 ER -