Abstract
The significant power dissipation of the on-chip L2 cache is a major concern in modern microprocessors. This Letter proposes a unique low-power technique for a high-associative large L2 cache that has fragmented locality by L1 cache. The dynamic per history length adjustment cache (DHL-cache) dynamically selects the qualified way candidates to be accessed and controls the way-prediction window size based on the access history pattern. With a high degree of the way-prediction accuracy, the DHL-cache shows a 55.3% energy-delay product improvement over the location-cache with minimum hardware support. Therefore, the DHL-cache alleviates the power limit issue for L2 cache, even with high associativity and fragmented locality.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
