Disperse Access Considered Energy Inefficiency in Intel Optane DC Persistent Memory Servers
Disperse Access Considered Energy Inefficiency in Intel Optane DC Persistent Memory Servers
复制标题
DOI：
10.1109/icdcs47774.2020.00107
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Daping Li;Ji-guang Wan;Jun Wang;Jian Zhou;Kai Lu;Peng Xu;Fei Wu;C. Xie
中科院分区：
文献类型：
--
作者：
Daping Li;Ji-guang Wan;Jun Wang;Jian Zhou;Kai Lu;Peng Xu;Fei Wu;C. Xie
Intel Optane DC Persistent Memory Module（AEP）是第一款商用非易失性存储器（NVM）产品，可提供与DRAM相当的性能，同时提供更大的容量和数据持久性。现有的用DRAM替代NVM或混合它们的研究要么是基于仿真器的，要么集中在如何提高写入的能量效率。不幸的是，真实的AEP系统的能量效率较少被探索。基于真实的AEP，我们观察到，即使消除了DRAM类刷新能耗，AEP在不同的性能水平上消耗的能量也有很大的不同。具体而言，与没有时间间隔的请求（紧凑）的情况相比，具有时间间隔的请求（分散）在性能和能源效率方面表现不佳。这种差异和并行开发潜力促使我们提出Sprint-AEP，一种面向能效的AEP服务器调度方法。Sprint-AEP通过延迟写请求和预取最热的数据来完全激活足够的AEP以服务于大多数请求。剩余的AEP将保持在具有低空闲功率的空闲模式以节省能量。此外，我们还利用读并行来加速同步和预取过程。实验结果表明，与能量未知的AEP使用相比，Sprint-AEP节省了高达26%的能量，性能几乎没有下降。
The Intel Optane DC Persistent Memory Module (AEP), which is the first commercial available Non-Volatile Memory (NVM) product, offers comparable performance with DRAM while providing larger capacities and data persistence. Existing researches that substitute NVM with DRAM or hybridize them are either emulator-based or focused on how to improve the energy efficiency for writes. Unfortunately, the energy efficiency of the real AEP system is less explored. Based on real AEP, we observe that even though eliminating the DRAM-like refresh energy consumptions, AEP consumes significant different energy at different performance levels. Specifically, requests with time intervals (dispersed) underperform in both performance and energy efficiency when compared with the case of requests without time intervals (compact). This disparity and parallelism exploitation potentials motivate us to propose Sprint-AEP, an energy-efficiency-oriented scheduling method for AEP-equipped servers. Sprint-AEP fully activates adequate AEPs to serve most of the requests by deferring the write requests and prefetching the hottest data. The remaining AEPs will stay in idle mode with a low idle power to save energy. Besides, we also utilize the read parallelism to accelerate the sync and prefetching processes. Compared with energy-unaware AEP usages, our experimental results show that Sprint-AEP saves up to 26% energy with little performance degradation.