SSCMSD - Single-Symbol Correction Multi-symbol Detection for DRAM Subsystem

Ravikiran Yeleswarapu, Arun Kumar Somani
{"title":"SSCMSD - Single-Symbol Correction Multi-symbol Detection for DRAM Subsystem","authors":"Ravikiran Yeleswarapu, Arun Kumar Somani","doi":"10.1109/PRDC.2018.00012","DOIUrl":null,"url":null,"abstract":"As DRAM technology continues to evolve towards smaller feature sizes and increased densities, faults in DRAM subsystem are becoming more severe. Current servers mostly use CHIPKILL based schemes to tolerate up-to one/two symbol errors per DRAM beat. Multi-symbol errors arising due to faults in multiple data buses and chips may not be detected by these schemes. In this paper, we introduce Single Symbol Correction Multiple Symbol Detection (SSCMSD) - a novel error handling scheme to correct single-symbol errors and detect multi-symbol ones. Here, we use a hash in combination with ECC to avoid silent data corruptions (SDCs). We employ 32-bit Spookyhash along with Reed-Solomon code to implement SSCMSD for a x4 based DDRx system. Our simulations show that the proposed scheme effectively prevents SDCs in the presence of multiple symbol errors. For this design, we need 19 chips per rank (storage overhead of 18.75 percent), 76 data bus-lines and additional hash-logic at the memory controller.","PeriodicalId":409301,"journal":{"name":"2018 IEEE 23rd Pacific Rim International Symposium on Dependable Computing (PRDC)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE 23rd Pacific Rim International Symposium on Dependable Computing (PRDC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PRDC.2018.00012","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

Abstract

As DRAM technology continues to evolve towards smaller feature sizes and increased densities, faults in DRAM subsystem are becoming more severe. Current servers mostly use CHIPKILL based schemes to tolerate up-to one/two symbol errors per DRAM beat. Multi-symbol errors arising due to faults in multiple data buses and chips may not be detected by these schemes. In this paper, we introduce Single Symbol Correction Multiple Symbol Detection (SSCMSD) - a novel error handling scheme to correct single-symbol errors and detect multi-symbol ones. Here, we use a hash in combination with ECC to avoid silent data corruptions (SDCs). We employ 32-bit Spookyhash along with Reed-Solomon code to implement SSCMSD for a x4 based DDRx system. Our simulations show that the proposed scheme effectively prevents SDCs in the presence of multiple symbol errors. For this design, we need 19 chips per rank (storage overhead of 18.75 percent), 76 data bus-lines and additional hash-logic at the memory controller.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
DRAM子系统的单符号校正多符号检测
随着DRAM技术不断向更小的特征尺寸和更高的密度发展,DRAM子系统的故障也越来越严重。当前的服务器大多使用基于CHIPKILL的方案来容忍每个DRAM节拍最多一/两个符号错误。这些方案可能无法检测到由多个数据总线和芯片故障引起的多符号错误。本文介绍了单符号校正多符号检测(SSCMSD)——一种新的错误处理方案,用于校正单符号错误和检测多符号错误。在这里,我们将散列与ECC结合使用,以避免静默数据损坏(sdc)。我们使用32位Spookyhash和Reed-Solomon代码为基于x4的DDRx系统实现SSCMSD。仿真结果表明,该方案可以有效地防止存在多个符号错误的sdc。对于这个设计,我们需要每个等级19个芯片(存储开销为18.75%)、76条数据总线和内存控制器上的额外哈希逻辑。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
SSCMSD - Single-Symbol Correction Multi-symbol Detection for DRAM Subsystem Cyber-Physical Transactions: A Method for Securing VANETs with Blockchains An Approach for Trustworthiness Benchmarking Using Software Metrics Semantic Failover in Software-Defined Networking InfoLeak: Scheduling-Based Information Leakage
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1