research-article

Revisiting widely held SSD expectations and rethinking system-level implications

Authors:

Myoungsoo Jung,

Mahmut KandemirAuthors Info & Claims

ACM SIGMETRICS Performance Evaluation Review, Volume 41, Issue 1

Pages 203 - 216

https://doi.org/10.1145/2494232.2465548

Published: 17 June 2013 Publication History

Abstract

Storage applications leveraging Solid State Disk (SSD) technology are being widely deployed in diverse computing systems. These applications accelerate system performance by exploiting several SSD-specific characteristics. However, modern SSDs have undergone a dramatic technology and architecture shift in the past few years, which makes widely held assumptions and expectations regarding them highly questionable. The main goal of this paper is to question popular assumptions and expectations regarding SSDs through an extensive experimental analysis using 6 state-of-the-art SSDs from different vendors. Our analysis leads to several conclusions which are either not reported in prior SSD literature, or contradict to current conceptions. For example, we found that SSDs are not biased toward read-intensive workloads in terms of performance and reliability. Specifically, random read performance of SSDs is worse than sequential and random write performance by 40% and 39% on average, and more importantly, the performance of sequential reads gets significantly worse over time. Further, we found that reads can shorten SSD lifetime more than writes, which is very unfortunate, given the fact that many existing systems/platforms already employ SSDs as read caches or in applications that are highly read intensive. We also performed a comprehensive study to understand the worst-case performance characteristics of our SSDs, and investigated the viability of recently proposed enhancements that are geared towards alleviating the worst-case performance challenges, such as TRIM commands and background-tasks. Lastly, we uncover the overheads of these enhancements and their limits, and discuss system-level implications.

References

[1]

Arpaci-Dusseau, A. C., et al. Removing the costs of indirection in flash-based ssds with namelesswrites. In HotStorage (2010).

Digital Library

[2]

Cai, Y., et al. Flash correct-and-refresh: Retention-aware error management. In ICCD (2012).

[3]

Canim, M., et al. SSD bufferpool extensions for database systems. VLDB (2010).

Digital Library

[4]

Caulfield, A. M., et al. Gordon: Using flash memory to build fast, power-efficient clusters for data-intensive applications. In ASPLOS (2009).

Digital Library

[5]

Caulfield, A. M., et al. Moneta: A high-performance storage array architecture for next-generation, non-volatile memories. In MICRO (2010).

Digital Library

[6]

Chen, F., et al. Essential roles of exploiting internal parallelism of flash memory based solid state drives in high-speed data processing. In HPCA (2011).

Digital Library

[7]

Choi, H., et al. VLSI implementation of BCH error correction for multilevel cell nand flash memory. In VLSI (2010).

Digital Library

[8]

Choudhuri, S., and Givargis, T. Deterministic service guarantees for NAND flash using partial block cleaning. In CODES+ISSS (2008).

Digital Library

[9]

Cooke, J. How ClearNAND flash simplifies and enhances system designs. In Micron White Paper (2011).

[10]

Dirik, C., and Jacob, B. The performance of PC solid-state disks (SSDs) as a function of bandwidth, concurrency, device architecture, and system organization. In ISCA (2009).

Digital Library

[11]

Feng Chen and others. Understanding intrinsic characteristics and system implications of flash memory based solid state drives. In SIGMETRICS (2009).

Digital Library

[12]

Fusion-Io. ioCache. In datasheet (2012).

[13]

Fusion-Io. ioMemory. In datasheet (2012).

[14]

Fusion-Io. ioTurbine. In datasheet (2012).

[15]

Gupta, A., et al. DFTL: A flash translation layer employing demand-based selective caching of page-level address mappings. In ASPLOS (2009).

Digital Library

[16]

Hu, X.-Y., et al. Write amplification analysis in flash-based solid state drives. In SYSTOR (2009).

Digital Library

[17]

Hu, Y., et al. Performance impact and interplay of SSD parallelism through advanced commands, allocation strategy and data granularity. In ISC (2011).

Digital Library

[18]

Intel. http://www.iometer.org/. In Iometer User's Guide (2003), Intel.

[19]

J. Kang et al. A superblock-based flash translation layer for NAND flash memory. In EMSOFT (2006).

Digital Library

[20]

Jo, H., et al. FAB: flash-aware buffer management policy for portable media players.

[21]

Josephson, W. K., et al. Dfs: A file system for virtualized flash storage. In FAST (2010).

Digital Library

[22]

Jung, M., et al. Memory system and data storing method thereof. U.S. Patent 20090248987 (2009).

[23]

Jung, M., et al. Physically addressed queueing (PAQ): Improving parallelism in solid state disks. In ISCA (2012).

Digital Library

[24]

Jung, M., et al. Taking garbage collection overheads off the critical path in ssds. In Middleware (2012).

Digital Library

[25]

Jung, M., and Kandemir, M. An evaluation of different page allocation strategies on high-speed SSDs. In HotStorage (2012).

Digital Library

[26]

Jung, M., and Kandemir, M. Middleware - firmware cooperation for high-speed solid state drives. In Middleware D&P (2012).

Digital Library

[27]

Jung, M., and Yoo, J. Scheduling garbage collection opportunistically to reduce worst-case I/O performance in solid state disks. In IWSSPS (2009).

[28]

Kgil, T., Roberts, D., and Mudge, T. Improving NAND flash based disk caches. In ISCA (2008).

Digital Library

[29]

Kim, H., and Ahn, S. BPLRU: A buffer management scheme for improving random writes in flash storage. In FAST (2008).

Digital Library

[30]

Koltsidas, I., and Viglas, S. The case for flash-aware multi level caching.

[31]

LeCroy. http://www.lecroy.com/.

[32]

Liu, N., et al. On the role of burst buffers in leadership-class storage systems. In MSST (2012).

[33]

Liu, Y., Huang, J., Xie, C., and Cao, Q. Raf: A random access first cache management to improve SSD-based disk cache. NAS (2010).

Digital Library

[34]

N. Agrawal et al. Design tradeoffs for SSD performance. In USENIX ATC (2008).

Digital Library

[35]

ONFI Working Group. Open nand flash interface 3.0. In http://onfi.org/ (2012).

[36]

Ou, Y., et al. Cfdc: a flash-aware replacement policy for database buffer management. In DAMON (2009).

Digital Library

[37]

Ouyang, X., et al. Enhancing checkpoint performance with staging I/O and SSD. In SNAPI (2010).

Digital Library

[38]

Saxena, M., et al. FlashVM: Virtual memory management on flash. In USENIX ATC (2010).

Digital Library

[39]

Srinivasan, M., and Callaghan, M. Flashcache at facebook. In Facebook White Paper (2010).

[40]

T13. Serial ATA Specification 3.1. 2012.

[41]

Zhang, Y., et al. De-indirection for flash-based ssds with namelesswrites. In FAST (2012).

Digital Library

Cited By

Jang HJin YLee DLee B(2024)Dynamic Clustering Page Allocation for Read-Intensive Multimedia Streaming Applications2024 IEEE 21st Consumer Communications & Networking Conference (CCNC)10.1109/CCNC51664.2024.10454852(84-89)Online publication date: 6-Jan-2024
https://doi.org/10.1109/CCNC51664.2024.10454852
Park JEom Y(2023)Filesystem Fragmentation on Modern Storage SystemsACM Transactions on Computer Systems10.1145/361138641:1-4(1-27)Online publication date: 18-Dec-2023
https://dl.acm.org/doi/10.1145/3611386
Kwon MLee SChoi HHwang JJung M(2023)Realizing Strong Determinism Contract on Log-Structured Merge Key-Value StoresACM Transactions on Storage10.1145/358269519:2(1-29)Online publication date: 25-Mar-2023
https://dl.acm.org/doi/10.1145/3582695
Show More Cited By

Index Terms

Revisiting widely held SSD expectations and rethinking system-level implications

Recommendations

Revisiting widely held SSD expectations and rethinking system-level implications
SIGMETRICS '13: Proceedings of the ACM SIGMETRICS/international conference on Measurement and modeling of computer systems

Storage applications leveraging Solid State Disk (SSD) technology are being widely deployed in diverse computing systems. These applications accelerate system performance by exploiting several SSD-specific characteristics. However, modern SSDs have ...
An efficient design and implementation of LSM-tree based key-value store on open-channel SSD
EuroSys '14: Proceedings of the Ninth European Conference on Computer Systems

Various key-value (KV) stores are widely employed for data management to support Internet services as they offer higher efficiency, scalability, and availability than relational database systems. The log-structured merge tree (LSM-tree) based KV stores ...
RAID-Aware SSD: Improving the Write Performance and Lifespan of SSD in SSD-Based RAID-5 System
BDCLOUD '14: Proceedings of the 2014 IEEE Fourth International Conference on Big Data and Cloud Computing

Flash memory-based SSD RAID has an excellent I/O performance with high stability, which making it get more and more attention from companies and manufacturers, especially in I/O-intensive environments. However, frequently updating parity also makes the ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image ACM SIGMETRICS Performance Evaluation Review

ACM SIGMETRICS Performance Evaluation Review Volume 41, Issue 1

Performance evaluation review

June 2013

385 pages

ISSN:0163-5999

DOI:10.1145/2494232

Issue’s Table of Contents

SIGMETRICS '13: Proceedings of the ACM SIGMETRICS/international conference on Measurement and modeling of computer systems
June 2013
406 pages
ISBN:9781450319003
DOI:10.1145/2465529
General Chair:
Mor Harchol-Balter
Carnegie Mellon University, USA
,
Program Chairs:
John Douceur
Microsoft Research, USA
,
Jun Xu
Georgia Institute of Technology, USA

Copyright © 2013 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 17 June 2013

Published in SIGMETRICS Volume 41, Issue 1

Check for updates

Author Tags

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

138
Total Citations
View Citations
1,300
Total Downloads

Downloads (Last 12 months)88
Downloads (Last 6 weeks)9

Reflects downloads up to 12 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Jang HJin YLee DLee B(2024)Dynamic Clustering Page Allocation for Read-Intensive Multimedia Streaming Applications2024 IEEE 21st Consumer Communications & Networking Conference (CCNC)10.1109/CCNC51664.2024.10454852(84-89)Online publication date: 6-Jan-2024
https://doi.org/10.1109/CCNC51664.2024.10454852
Park JEom Y(2023)Filesystem Fragmentation on Modern Storage SystemsACM Transactions on Computer Systems10.1145/361138641:1-4(1-27)Online publication date: 18-Dec-2023
https://dl.acm.org/doi/10.1145/3611386
Kwon MLee SChoi HHwang JJung M(2023)Realizing Strong Determinism Contract on Log-Structured Merge Key-Value StoresACM Transactions on Storage10.1145/358269519:2(1-29)Online publication date: 25-Mar-2023
https://dl.acm.org/doi/10.1145/3582695
Han SCho MLee GChung E(2023)Page Type-Aware Data Migration Technique for Read Disturb Management of NAND Flash MemoryIEEE Transactions on Very Large Scale Integration (VLSI) Systems10.1109/TVLSI.2023.324017231:4(591-595)Online publication date: 1-Apr-2023
https://dl.acm.org/doi/10.1109/TVLSI.2023.3240172
Park JEom YAnwar ASkourtis DKannan SMa X(2022)File fragmentation from the perspective of I/O controlProceedings of the 14th ACM Workshop on Hot Topics in Storage and File Systems10.1145/3538643.3539746(126-132)Online publication date: 27-Jun-2022
https://dl.acm.org/doi/10.1145/3538643.3539746
Bae HKim JKwon MJung MAnwar ASkourtis DKannan SMa X(2022)What you can't forgetProceedings of the 14th ACM Workshop on Hot Topics in Storage and File Systems10.1145/3538643.3539744(79-85)Online publication date: 27-Jun-2022
https://dl.acm.org/doi/10.1145/3538643.3539744
Bae JPark JJun YSeo EMalka MKolodner HBellosa FGabel M(2022)Dedup-for-speedProceedings of the 15th ACM International Conference on Systems and Storage10.1145/3534056.3534937(128-139)Online publication date: 6-Jun-2022
https://dl.acm.org/doi/10.1145/3534056.3534937
Feng XChen XLi RLi JSong CLiu DTan YQiao L(2022)CoDiscardJournal of Systems Architecture: the EUROMICRO Journal10.1016/j.sysarc.2022.102564128:COnline publication date: 1-Jul-2022
https://dl.acm.org/doi/10.1016/j.sysarc.2022.102564
Lee GShin SJeong J(2022)Efficient hybrid polling for ultra-low latency storage devicesJournal of Systems Architecture: the EUROMICRO Journal10.1016/j.sysarc.2021.102338122:COnline publication date: 1-Jan-2022
https://dl.acm.org/doi/10.1016/j.sysarc.2021.102338
Dayan NTwitto MRochman YBeitler UZion IBortnikov EDashevsky SFrishman OGinzburg EMaly IMeir AMokryn MNaiss IRabinovich N(2021)The end of Moore's law and the rise of the data processorProceedings of the VLDB Endowment10.14778/3476311.347637314:12(2932-2944)Online publication date: 28-Oct-2021
https://dl.acm.org/doi/10.14778/3476311.3476373
Show More Cited By

View Options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Issue’s Table of Contents