Idler : I/O Workload Controlling for Better Responsiveness on Host-Aware Shingled Magnetic Recording Drives

Host-Aware/Drive-Managed Shingled Magnetic Recording (SMR) drives can accept non-sequential writes using a buffer called media cache. Data in the media cache will be migrated to its designated location by a cleaning process if the buffer is full (blocking cleaning) or the drive is idle (idle cleaning). However, blocking cleanings can severely extend the I/O response time. Therefore, it is crucial to fully understand the cleaning process and find ways of mitigating the caused performance degradation. In this article we further evaluate the cleaning process and propose a potential remedy scheme called Idler on Host-Aware SMR drives. Idler adaptively induces idle cleanings based on dynamic workload characteristics and media cache usages to reduce the severity of blocking cleanings. Our evaluations show that in the workloads with a small non-sequential write ratio (about 10 percent), Idler can reduce the tail response time and the workload finish time by 56–88 and 10–23 percent, respectively, compared with those without such control. With the help of an external write buffer on an SSD, the tail response time of SMR drives with Idler can be closer to that of conventional disk drives.

[1]  David Hung-Chang Du,et al.  H-SWD: Incorporating Hot Data Identification into Shingled Write Disks , 2012, 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[2]  Ting Yao,et al.  GearDB: A GC-free Key-Value Store on HM-SMR Drives with Gear Compaction , 2019, FAST.

[3]  Abutalib Aghayev,et al.  Evolving Ext4 for Shingled Disks , 2017, FAST.

[4]  Ziqi Fan,et al.  A Lookahead Read Cache: Improving Read Performance for Deduplication Backup Storage , 2017, Journal of Computer Science and Technology.

[5]  Andrew A. Chien,et al.  The Tail at Store: A Revelation from Millions of Hours of Disk and SSD Deployments , 2016, FAST.

[6]  Tei-Wei Kuo,et al.  On Improving the Write Responsiveness for Host-Aware SMR Drives , 2019, IEEE Transactions on Computers.

[7]  P. Desnoyers,et al.  Skylight—A Window on Shingled Disk Operation , 2015, FAST.

[8]  Garth A. Gibson,et al.  Shingled Magnetic Recording: Areal Density Increase Requires New Data Management , 2013, login Usenix Mag..

[9]  Paul Rad,et al.  Designing and evaluating hybrid storage for high performance cloud computing , 2018, 2018 Annual IEEE International Systems Conference (SysCon).

[10]  Dandan Wang,et al.  Larger , Cheaper , but Faster : SSD-SMR Hybrid Storage Boosted by a New SMR-oriented Cache Framework , 2017 .

[11]  David Hung-Chang Du,et al.  SMaRT: An Approach to Shingled Magnetic Recording Translation , 2017, FAST.

[12]  Yuan-Hao Chang,et al.  Improving Runtime Performance of Deduplication System with Host-Managed SMR Storage Drives , 2018, 2018 55th ACM/ESDA/IEEE Design Automation Conference (DAC).

[13]  Ethan L. Miller,et al.  SMRDB: key-value data store for shingled magnetic recording disks , 2015, SYSTOR.

[14]  Jialin Li,et al.  Tales of the Tail: Hardware, OS, and Application-level Sources of Tail Latency , 2014, SoCC.

[15]  James Kelley,et al.  SMORE: A Cold Data Object Store for SMR Drives (Extended Version) , 2017, ArXiv.

[16]  Seung-won Hwang,et al.  Delayed-Dynamic-Selective (DDS) Prediction for Reducing Extreme Tail Latency in Web Search , 2015, WSDM.

[17]  Antony I. T. Rowstron,et al.  Write off-loading: Practical power management for enterprise storage , 2008, TOS.

[18]  Mor Harchol-Balter,et al.  PriorityMeister: Tail Latency QoS for Shared Networked Storage , 2014, SoCC.

[19]  Abutalib Aghayev,et al.  Modeling SMR Drive Performance , 2016, SIGMETRICS.

[20]  Ziqi Fan,et al.  Evaluating Host Aware SMR Drives , 2016, HotStorage.

[21]  Dan Feng,et al.  PSLO: enforcing the Xth percentile latency and throughput SLOs for consolidated VM storage , 2016, EuroSys.

[22]  David Hung-Chang Du,et al.  On the Accuracy and Scalability of Intensive I/O Workload Replay , 2017, FAST.

[23]  Carlos Maltzahn,et al.  ZEA, A Data Management Approach for SMR , 2016, HotStorage.

[24]  Liquan Xiao,et al.  SMRC: An Endurable SSD Cache for Host-Aware Shingled Magnetic Recording Drives , 2018, IEEE Access.

[25]  Chao Jin,et al.  HiSMRfs: A high performance file system for shingled storage array , 2014, 2014 30th Symposium on Mass Storage Systems and Technologies (MSST).

[26]  Ricardo Bianchini,et al.  Few-to-Many: Incremental Parallelism for Reducing Tail Latency in Interactive Services , 2015, ASPLOS.