Shared Last-Level Caches and The Case for Longer Timeslices Conference Paper uri icon

abstract

  • Memory performance is important in modern systems. Contention at various levels in memory hierarchy can lead to signifcant application performance degradation due to interference. Further, modern, large, last-level caches (LLC) have fill times greater than the OS scheduling window. When several threads are running concurrently and timesharing the CPU cores, they may never be able to load their working sets into the cache before being rescheduled, thus permanently stuck in the "cold-start" regime. We show that by increasing the system scheduling timeslice length it is possible to amortize the cache cold-start penalty due to the multitasking and improve application performance by 10{ 15%.

name of conference

  • Proceedings of the 2015 International Symposium on Memory Systems

published proceedings

  • Proceedings of the 2015 International Symposium on Memory Systems

author list (cited authors)

  • Fedorov, V. V., Reddy, A., & Gratz, P. V.

citation count

  • 5

complete list of authors

  • Fedorov, Viacheslav V||Reddy, AL Narasimha||Gratz, Paul V

editor list (cited editors)

  • Jacob, B. L.

publication date

  • October 2015