2 Classifying Data to Reduce Long-Term Data Movement in Shingled Write Disks STEPHANIE N. JONES, University of California, Santa Cruz AHMED AMER, Santa Clara University ETHAN L. MILLER, DARRELL D. E. LONG, REKHA PITCHUMANI, and CHRISTINA R. STRONG, University of California, Santa Cruz Shingled magnetic recording (SMR) is a means of increasing the density of hard drives that brings a new set of challenges. Due to the nature of SMR disks, updating in place is not an option. Holes left by invalidated data can only be filled if the entire band is reclaimed, and a poor band compaction algorithm could result in spending a lot of time moving blocks over the lifetime of the device. We propose using write frequency to separate blocks to reduce data movement and develop a band compaction algorithm that implements this heuristic. We demonstrate how our algorithm results in improved data management, resulting in an up to 45% reduction in required data movements when compared to naive approaches to band management. CCS Concepts:  Information systems → Magnetic disks;  Hardware → Memory and dense stor- age;  Computing methodologies → Modeling and simulation Additional Key Words and Phrases: Storage, shingled write disks, shingled magnetic recording drives, data placement ACM Reference Format: Stephanie N. Jones, Ahmed Amer, Ethan L. Miller, Darrell D. E. Long, Rekha Pitchumani, and Christina R. Strong. 2016. Classifying data to reduce long-term data movement in shingled write disks. ACM Trans. Storage 12, 1, Article 2 (February 2016), 17 pages. DOI: http://dx.doi.org/10.1145/2851505 1. INTRODUCTION Shingled magnetic recording (SMR) disks are devices that increase the storage den- sity of traditional disk media by writing overlapping wide tracks of data, resulting in a shingle-like arrangement of the tracks. This presents us with a problem when overwriting previously written data, as it cannot be done without overwriting adjacent tracks. Overlapping tracks are therefore arranged into bands, and space reclamation is done at a band level. SMR disks therefore must employ band compaction to reclaim bands containing overwritten data, a concept similar to log-structured file system (LFS) cleaning. A This work is supported in part by the National Science Foundation under award IIP-1266400 and industrial members of the Center for Research in Storage Systems. A. Amer and D. D. E. Long were supported in part by the National Science Foundation under awards CCF-1219163 and CCF-1217648, by the Department of Energy under award DE-FC02-10ER26017/DE-SC0005417, by the industrial members of the Storage Systems Research Center, and by a gift from Wells Fargo. Authors’ addresses: S. N. Jones, A. Amer, E. L. Miller, D. D. E. Long, R. Pitchumani, and C. R. Strong, 1156 High Street, Santa Cruz, CA 95064. Permission to make digital or hard copies of part or all 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 show this notice on the first page or initial screen of a display along with the full citation. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, to redistribute to lists, or to use any component of this work in other works requires prior specific permission and/or a fee. Permissions may be requested from Publications Dept., ACM, Inc., 2 Penn Plaza, Suite 701, New York, NY 10121-0701 USA, fax +1 (212) 869-0481, or permissions@acm.org. c  2016 ACM 1553-3077/2016/02-ART2 $15.00 DOI: http://dx.doi.org/10.1145/2851505 ACM Transactions on Storage, Vol. 12, No. 1, Article 2, Publication date: February 2016.