Sanitizer: Mitigating the Impact of Expensive ECC Checks on STT-MRAM Based Main Memories
Sanitizer: Mitigating the Impact of Expensive ECC Checks on STT-MRAM Based Main Memories
复制标题
Sanitizer：减轻昂贵的 ECC 检查对基于 STT-MRAM 的主存储器的影响
DOI：
--
复制
发表时间：
2018
影响因子：
3.7
通讯作者：
Engin Ipek
中科院分区：
文献类型：
--
作者：
Xiaochen Guo;M. N. Bojnordi;Qing Guo;Engin Ipek
由于在纳米级特征尺寸下保持足够高的存储电容和足够低的泄漏电流的挑战，DRAM密度缩放变得越来越困难。非易失性存储器（NVMs）作为潜在的DRAM替代品引起了极大的关注，因为它们使用电阻而不是电荷来表示信息。自旋转矩传递磁阻存储器（STT-MRAM）由于其相对低的写入能量、高的速度和高的耐用性，是最有前途的NVM技术之一。然而，STT-MRAM也有其自身的规模问题。随着存储元件的尺寸随着技术的扩展而减小，STT-MRAM保留错误率预计会增加，这将需要多比特纠错码（ECC）和定期擦洗。我们介绍了<italic>Sanitizer</italic>架构，该架构降低了ECC和在未来基于STT-MRAM的主存储器中擦洗的性能和能源开销。为了降低刷洗率，使用了一种具有12.5%存储开销的粗粒度多比特ECC机制。为了避免从内存中获取多个块并在每次读取时执行昂贵的ECC检查，可以预测并主动清除在不久的将来可能访问的内存区域。与传统的STT-MRAM系统相比，Sanitizer将性能提高了1.22<inline-formula> < text -math notation="LaTeX">$ times $</ text -math><alternatives><inline-graphic xlink:href=" gu- ieq1-2779151.gif"/> </alternatives></inline-formula>，并将端到端系统能耗降低了22%。
DRAM density scaling has become increasingly difficult due to challenges in maintaining a sufficiently high storage capacitance and a sufficiently low leakage current at nanoscale feature sizes. Non-volatile memories (NVMs) have drawn significant attention as potential DRAM replacements because they represent information using resistance rather than electrical charge. Spin-torque transfer magnetoresistive RAM (STT-MRAM) is one of the most promising NVM technologies due to its relatively low write energy, high speed, and high endurance. However, STT-MRAM suffers from its own scaling problems. As the size of the storage element decreases with technology scaling, STT-MRAM retention error rates are expected to increase, which will require multi-bit error-correcting code (ECC) and periodic scrubbing. We introduce the <italic>Sanitizer</italic> architecture, which mitigates the performance and energy overheads of ECC and scrubbing in future STT-MRAM based main memories. To reduce the scrubbing rate, a coarse-grained, multi-bit ECC mechanism with a 12.5 percent storage overhead is used. To avoid fetching multiple blocks from memory and performing costly ECC checks on every read, the memory regions that will likely be accessed in the near future are predicted and proactively scrubbed. Compared to a conventional STT-MRAM system, Sanitizer improves performance by 1.22<inline-formula> <tex-math notation="LaTeX">$ imes$</tex-math><alternatives><inline-graphic xlink:href="guo-ieq1-2779151.gif"/> </alternatives></inline-formula> and reduces end-to-end system energy by 22 percent.