教師資料查詢 | 類別: 會議論文 | 教師: 衛信文 Wei, Hsin-Wen (瀏覽個人網頁)

標題:Reinforcing the energy efficiency of cyber-physical systems via direct and split cache consolidation on MLC STT-RAM
學年108
學期2
發表日期2020/03/30
作品名稱Reinforcing the energy efficiency of cyber-physical systems via direct and split cache consolidation on MLC STT-RAM
作品名稱(其他語言)
著者Shuo-Han Chen; Yu-Pei Liang; Yuan-Hao Chang; Yun-Fei Liu; Chun-Feng Wu; Hsin-Wen Wei; Wei-Kuan Shih
作品所屬單位
出版者
會議名稱SAC '20: The 35th ACM/SIGAPP Symposium on Applied Computing
會議地點online event, [Brno, Czech Republic]
摘要Energy efficiency has become one of the primary considerations in the designs of cyber-physical systems (CPS). However, CPS with static RAM (SRAM)-based processors suffers from the high leakage power issue of SRAM, thus limiting the energy efficiency of CPS. Recently, Spin-Transfer Torque RAM (STT-RAM) has emerged and been widely regarded as a great alternative as the on-chip memory within processors, owing to STT-RAM's high density and near-zero leakage power characteristics. In addition, recent advances in Magnetic Tunneling Junction (MTJ) technology also realize the multi-level cell (MLC) STT-RAM to further enhance the memory density. Nevertheless, the write disturbance issue greatly limits the energy efficiency of MLC STT-RAM. Even though studies have been proposed to alleviate this issue, most of the previous disturbance reduction strategies could induce additional management overhead by utilizing counters or cause frequent swap operations when the write disturbance happens. Such observations motivate us to propose a simple and effective solution to unify the direct and split mapping cache designs for improving the energy efficiency of MLC STT-RAM. The proposed design is evaluated through a series of experiments on an emulator with encouraging results.
關鍵字cyber-physical systems;STT-RAM;MLC STT-RAM;direct mapping;cell split mapping
語言英文(美國)
收錄於
會議性質國際
校內研討會地點
研討會時間20200330~20200403
通訊作者
國別捷克
公開徵稿
出版型式
出處SAC '20: Proceedings of the 35th Annual ACM Symposium on Applied Computing, p.202–209
相關連結
Google+ 推薦功能,讓全世界都能看到您的推薦!