JZUS - Journal of Zhejiang University SCIENCE

ENGINEERING Information Technology & Electronic Engineering

Accepted manuscript available online (unedited version)

A novel non-volatile memory storage system for I/O-intensive applications

Author(s): Wen-bing Han, Xiao-gang Chen, Shun-fen Li, Ge-zi Li, Zhi-tang Song, Da-gang Li, Shi-yan Chen
Affiliation(s): Shanghai Institute of Micro-system and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China; more
Corresponding email(s): hwbx@mail.sim.ac.cn, chenxg@mail.sim.ac.cn
Key Words: In-storage processing, File system, Non-volatile memory (NVM), Storage system, I/O-intensive applications

Share this article to： More <<< Previous Paper \|

Wen-bing Han, Xiao-gang Chen, Shun-fen Li, Ge-zi Li, Zhi-tang Song, Da-gang Li, Shi-yan Chen. A novel non-volatile memory storage system for I/O-intensive applications[J]. Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/FITEE.1700061

@article{title="A novel non-volatile memory storage system for I/O-intensive applications",
author="Wen-bing Han, Xiao-gang Chen, Shun-fen Li, Ge-zi Li, Zhi-tang Song, Da-gang Li, Shi-yan Chen",
journal="Frontiers of Information Technology & Electronic Engineering",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/FITEE.1700061"
}

%0 Journal Article
%T A novel non-volatile memory storage system for I/O-intensive applications
%A Wen-bing Han
%A Xiao-gang Chen
%A Shun-fen Li
%A Ge-zi Li
%A Zhi-tang Song
%A Da-gang Li
%A Shi-yan Chen
%J Frontiers of Information Technology & Electronic Engineering
%P 1291-1302
%@ 2095-9184
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/FITEE.1700061"

TY - JOUR
T1 - A novel non-volatile memory storage system for I/O-intensive applications
A1 - Wen-bing Han
A1 - Xiao-gang Chen
A1 - Shun-fen Li
A1 - Ge-zi Li
A1 - Zhi-tang Song
A1 - Da-gang Li
A1 - Shi-yan Chen
J0 - Frontiers of Information Technology & Electronic Engineering
SP - 1291
EP - 1302
%@ 2095-9184
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/FITEE.1700061"

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: The emerging memory technologies, such as phase change memory (PCM), provide chances for high-performance storage of i/O-intensive applications. However, traditional software stack and hardware architecture need to be optimized to enhance I/O efficiency. In addition, narrowing the distance between computation and storage reduces the number of I/O requests and has become a popular research direction. This paper presents a novel PCM-based storage system. It consists of the in-storage processing enabled file system (ISPFS) and the configurable parallel computation fabric in storage, which is called an in-storage processing (ISP) engine. On one hand, ISPFS takes full advantage of non-volatile memory (NVM)’s characteristics, and reduces software overhead and data copies to provide low-latency high-performance random access. On the other hand, ISPFS passes ISP instructions through a command file and invokes the ISP engine to deal with I/O-intensive tasks. Extensive experiments are performed on the prototype system. The results indicate that ISPFS achieves 2 to 10 times throughput compared to EXT4. Our ISP solution also reduces the number of I/O requests by 97% and is 19 times more efficient than software implementation for i/O-intensive applications.

针对I/O密集型应用的新型非易失存储系统

摘要：新型存储技术—如相变存储器（phase change memory, PCM）的兴起，为I/O密集型应用的高性能存储提供了新机遇。然而传统软件栈和硬件架构需要重构优化以提升I/O访问效率。此外，存储内计算技术已经成为当前研究热点，能有效减少I/O请求数量。提出一种基于相变存储器的新型存储系统，包括支持存储内计算的文件系统（in-storage processing enabled filesy stem, ISPFS）和可配置并行计算的存储内计算引擎。一方面，ISPFS充分利用新型非易失存储器（non-volatile memory, NVM）的特性，减少软件冗余和数据拷贝，提供低延迟、高性能的随机访问。另一方面，ISPFS通过命令文件将ISP指令发送给存储内计算引擎来处理I/O密集型任务。实验证明，ISPFS的吞吐量是EXT4文件系统的2–10倍。对于I/O密集型应用，该存储内计算方案比传统软件方案减少97%的I/O请求，效率提升19倍。

关键词组：存储内计算；文件系统；非易失存储；存储系统；I/O密集型应用

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

[1]Axboe J, Brunelle AD, Scott N, 2006. blktrace(8) - Linux man page. https://linux.die.net/man/8/blktrace [Accessed on Jan. 19, 2016].

[2]Cao Q, 2012. SCMFS Performance Enhancement and Implementation on Mobile Platform. MS Thesis, Texas A&M University, College Station, Texas.

[3]Caulfield AM, Swanson S, 2013. QuickSAN: a storage area network for fast, distributed, solid state disks. Proc 40^th Annual Int Symp on Computer Architecture, p.464-474.

[4]Chen CLP, Zhang CY, 2014. Data-intensive applications, challenges, techniques and technologies: a survey on big data. Inform Sci, 275:314-347.

[5]Condit J, Nightingale EB, Frost C, et al., 2009. Better I/O through byte-addressable, persistent memory. Proc ACM SIGOPS 22^nd Symp on Operating Systems Principles, p.133-146.

[6]Do J, Kee YS, Patel JM, et al., 2013. Query processing on smart SSDs: opportunities and challenges. ACM SIGMOD Int Conf on Management of Data, p.1221-1230.

[7]Doller E, Akel A, Wang J, et al., 2014. DataCenter 2020: near-memory acceleration for data-oriented applications. Symp on VLSI Circuits Digest of Technical Papers, p.1-4.

[8]Han WB, Chen XG, Zhou M, et al., 2016. The storage system of PCM based on random access file system. Proc SPIE, 9818:98180G.

[9]Jun SW, Liu M, Fleming KE, et al., 2014. Scalable multi-access flash store for big data analytics. Proc ACM/SIGDA Int Symp on Field-Programmable Gate Arrays, p.55-64.

[10]Jun SW, Liu M, Lee S, et al., 2015. BlueDBM: an appliance for big data analytics. Proc ACM/IEEE 42^nd$ Annual Int Symp on Computer Architecture, p.1-13.

[11]Kang Y, Kee YS, Miller EL, et al., 2013. Enabling cost-effective data processing with smart SSD. Proc IEEE 29^th Symp on Mass Storage Systems and Technologies, p.1-12.

[12]Lee E, Bahn H, Yoo S, et al., 2014. Empirical study of NVM storage: an operating system’s perspective and implications. Proc IEEE 22^nd Int Symp on Modelling, Analysis & Simulation of Computer and Telecommunication Systems, p.405-410.

[13]Li GZ, Chen XG, Chen B, et al., 2016. An FPGA enhanced extensible and parallel query storage system for emerging NVRAM. IEICE Electron Expr, 13(4):20151109.

[14]Li Z, Wang F, Liu JN, et al., 2016. A user-visible solid-state storage system with software-defined fusion methods for PCM and NAND flash. J Syst Archit, 71:44-61.

[15]Norcott W, Capps D, 2016. IOZONE filesystem benchmark. http://www.iozone.org/ [Accessed on Jan. 23, 2016].

[16]Qiu S, Reddy ALN, 2013. NVMFS: a hybrid file system for improving random write in NAND-flash SSD. Proc IEEE 29^th Symp on Mass Storage Systems and Technologies, p.1-5.

[17]SAMSUNG, 2015. In-storage compute: an ultimate solution for accelerating I/O-intensive applications. http://www.flashmemorysummit.com/English/linebreak Collaterals/Proceedings/2015/20150813_S301Dlinebreak_Ki.pdf [Accessed on Dec. 11, 2016].

[18]Sha EHM, Chen XZ, Zhuge Q, et al., 2015. Designing an efficient persistent in-memory file system. IEEE Non-volatile Memory System and Applications Symp, p.1-6.

[19]Sha EHM, Chen XZ, Zhuge Q, et al., 2016. A new design of in-memory file system based on file virtual address framework. IEEE Trans Comput, 65(10):2959-2972.

[20]Szalay A, Gray J, 2006. 2020 computing: science in an exponential world. Nature, 440(7083):413-414.

[21]Wu XJ, Reddy ALN, 2011. SCMFS: a file system for storage class memory. Int Conf for High Performance Computing, Networking, Storage and Analysis, p.1-11.

[22]XILINX, 2014. Zynq-7000 all programmable SoC technical reference manual. https://www.xilinx.com/support/documentation/user_guides/ug585-Zynq-7000-TRM.pdf [Accessed on Dec. 11, 2016].

[23]Zhou M, Chen XG, Liu Y, et al., 2016. Design and implementation of a random access file system for NVRAM. IEICE Electron Expr, 13(4):20151045.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

- Go to

针对I/O密集型应用的新型非易失存储系统

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference