Single-pass restore after a media failure

When persistent storage fails, traditional media recovery first restores an old backup image followed by replaying the recovery log since the last backup operation. Restoring a backup can take hours, but log replay often takes much longer due to its random access pattern. We introduce single-pass restore, a technique in which restoration of all backups and log replay are performed in a single operation. This allows hiding log replay within the initial restore of the backup, thus substantially reducing the time and cost of media recovery and, incidentally, rendering incremental backup techniques unnecessary. Single-pass restore is enabled by a new organization of the log archive, created by a continuous process that is easily incorporated into the traditional log archiving process. Our empirical analysis shows that the imposed overhead is negligible in comparison with the substantial benefits provided.

[1]  Babak Falsafi,et al.  Shore-MT: a scalable storage manager for the multicore era , 2009, EDBT '09.

[2]  H KatzRandy,et al.  A case for redundant arrays of inexpensive disks (RAID) , 1988 .

[3]  Sandra Loosemore,et al.  The GNU C Library Reference Manual , 2001 .

[4]  Donald E. Knuth,et al.  The Art of Computer Programming: Volume 3: Sorting and Searching , 1998 .

[5]  C. Mohan,et al.  An efficient and flexible method for archiving a data base , 1993, SIGMOD Conference.

[6]  Harumi A. Kuno,et al.  Definition, Detection, and Recovery of Single-Page Failures, a Fourth Class of Database Failures , 2012, Proc. VLDB Endow..

[7]  Hamid Pirahesh,et al.  ARIES: a transaction recovery method supporting fine-granularity locking and partial rollbacks using write-ahead logging , 1998 .

[8]  Ippokratis Pandis,et al.  From A to E: analyzing TPC's OLTP benchmarks: the obsolete, the ubiquitous, the unexplored , 2013, EDBT '13.

[9]  Jim Gray,et al.  Notes on Data Base Operating Systems , 1978, Advanced Course: Operating Systems.

[10]  Andreas Reuter,et al.  Transaction Processing: Concepts and Techniques , 1992 .

[11]  Michael Stonebraker,et al.  Rethinking main memory OLTP recovery , 2014, 2014 IEEE 30th International Conference on Data Engineering.

[12]  Goetz Graefe,et al.  Implementing sorting in database systems , 2006, CSUR.

[13]  Randy H. Katz,et al.  A case for redundant arrays of inexpensive disks (RAID) , 1988, SIGMOD '88.

[14]  Goetz Graefe,et al.  An empirical analysis of database recovery costs , 2014 .

[15]  Hamid Pirahesh,et al.  ARIES-RRH: restricted repeating of history in the ARIES transaction recovery method , 1991, [1991] Proceedings. Seventh International Conference on Data Engineering.

[16]  Dina Bitton,et al.  Disk Shadowing , 1988, VLDB.