FragPicker: A New Defragmentation Tool for Modern Storage Devices

File fragmentation has been widely studied for several decades because it negatively influences various I/O activities. To eliminate fragmentation, most defragmentation tools migrate the entire content of files into a new area. Unfortunately, such methods inevitably generate a large amount of I/Os in the process of data migration. For this reason, the conventional tools (i) cause defragmentation to be time-consuming, (ii) significantly degrade the performance of co-running applications, and (iii) even curtail the lifetime of modern storage devices. Consequently, the current usage of defragmentation is very limited although it is necessary. Our extensive experiments discover that, unlike HDDs, the performance degradation of modern storage devices incurred by fragmentation mainly stems from request splitting, where a single I/O request is split into multiple ones. With this insight, we propose a new defragmentation tool, FragPicker, to minimize the amount of I/Os induced by defragmentation, while significantly improving I/O performance. FragPicker analyzes the I/O activities of applications and migrates only those pieces of data that are crucial to the I/O performance, in order to mitigate the aforementioned problems of existing tools. Experimental results demonstrate that FragPicker efficiently reduces the amount of I/Os for defragmentation while achieving a similar level of performance improvement to the conventional defragmentation schemes.

[1]  Hong Jiang,et al.  Improving Hybrid FTL by Fully Exploiting Internal SSD Parallelism with Virtual Blocks , 2014, ACM Trans. Archit. Code Optim..

[2]  洋一 中西,et al.  2012: , 2012, Disasters and Social Reproduction.

[3]  Kern Koh,et al.  A lifespan-aware reliability scheme for RAID-based flash storage , 2011, SAC '11.

[4]  Richard Taylor Interpretation of the Correlation Coefficient: A Basic Review , 1990 .

[5]  Nisha Talagala,et al.  HEC: improving endurance of high performance flash-based cache devices , 2013, SYSTOR '13.

[6]  Javier González,et al.  LightNVM: The Linux Open-Channel SSD Subsystem , 2017, FAST.

[7]  Young Ik Eom,et al.  Anti-Aging LFS: Self-Defragmentation With Fragmentation-Aware Cleaning , 2020, IEEE Access.

[8]  Ram Kesavan,et al.  Storage Gardening: Using a Virtualization Layer for Efficient Defragmentation in the WAFL File System , 2019, FAST.

[9]  Joo Young Hwang,et al.  F2FS: A New File System for Flash Storage , 2015, FAST.

[10]  Jack J. Dongarra,et al.  Exascale computing and big data , 2015, Commun. ACM.

[11]  David J. Lilja,et al.  Exploring Performance Characteristics of the Optane 3D Xpoint Storage Technology , 2020, ACM Trans. Model. Perform. Evaluation Comput. Syst..

[12]  Andrea C. Arpaci-Dusseau,et al.  Towards an Unwritten Contract of Intel Optane SSD , 2019, HotStorage.

[13]  Xavier Jimenez,et al.  Wear unleveling: improving NAND flash lifetime by balancing page endurance , 2014, FAST.

[14]  Evangelos Eleftheriou,et al.  Write amplification analysis in flash-based solid state drives , 2009, SYSTOR '09.

[15]  Gregory R. Ganger,et al.  Geriatrix: Aging what you see and what you don't see. A file system aging approach for modern storage systems , 2018, USENIX Annual Technical Conference.

[16]  Bryan Harris,et al.  Ultra-Low Latency SSDs' Impact on Overall Energy Efficiency , 2020, HotStorage.

[17]  J. R. Santos,et al.  Ext 4 block and inode allocator improvements , 2010 .

[18]  Andrea C. Arpaci-Dusseau,et al.  The Unwritten Contract of Solid State Drives , 2017, EuroSys.

[19]  Kai Liu,et al.  Boosting the Performance of SSDs via Fully Exploiting the Plane Level Parallelism , 2020, IEEE Transactions on Parallel and Distributed Systems.

[20]  Margo I. Seltzer,et al.  File system aging—increasing the relevance of file system benchmarks , 1997, SIGMETRICS '97.

[21]  Mahmut T. Kandemir,et al.  Revisiting widely held SSD expectations and rethinking system-level implications , 2013, SIGMETRICS '13.

[22]  Ram Kesavan,et al.  Countering Fragmentation in an Enterprise Storage System , 2020, ACM Trans. Storage.

[23]  Michael A. Bender,et al.  File Systems Fated for Senescence? Nonsense, Says Science! , 2017, FAST.

[24]  Fang Wang,et al.  ARS: Reducing F2FS Fragmentation for Smartphones using Decision Trees , 2020, 2020 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[25]  Chao Wu,et al.  An Empirical Study of File-System Fragmentation in Mobile Storage Systems , 2016, HotStorage.

[26]  Sungjin Lee,et al.  Lifetime improvement of NAND flash-based storage systems using dynamic program and erase scaling , 2014, FAST.

[27]  Hong Jiang,et al.  Exploring and Exploiting the Multilevel Parallelism Inside SSDs for Improved Performance and Endurance , 2013, IEEE Transactions on Computers.

[28]  Dan Williams,et al.  Platform Storage Performance With 3D XPoint Technology , 2017, Proceedings of the IEEE.

[29]  Sangwook Shane Hahn,et al.  File Fragmentation in Mobile Devices: Measurement, Evaluation, and Treatment , 2019, IEEE Transactions on Mobile Computing.

[30]  J. Brian Gray,et al.  Introduction to Linear Regression Analysis , 2002, Technometrics.

[31]  Yiming Hu,et al.  Parallelism and Garbage Collection Aware I/O Scheduler with Improved SSD Performance , 2017, 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS).

[32]  Michael A. Bender,et al.  Filesystem Aging: It's more Usage than Fullness , 2019, HotStorage.

[33]  Arif Merchant,et al.  Flash Reliability in Production: The Expected and the Unexpected , 2016, FAST.

[34]  Josef Bacik,et al.  BTRFS: The Linux B-Tree Filesystem , 2013, TOS.

[35]  Peter Mork,et al.  From Data to Decisions: A Value Chain for Big Data , 2013, IT Professional.

[36]  Xiaodong Zhang,et al.  Understanding intrinsic characteristics and system implications of flash memory based solid state drives , 2009, SIGMETRICS '09.

[37]  A. G. Asuero,et al.  The Correlation Coefficient: An Overview , 2006 .

[38]  Sungjin Lee,et al.  Improving File System Performance of Mobile Storage Systems Using a Decoupled Defragmenter , 2017, USENIX Annual Technical Conference.

[39]  Young Ik Eom,et al.  File Defragmentation Scheme for a Log-Structured File System , 2016, APSys.