Integrating reliable memory in databases

Abstract

Recent results in the Rio project at the University of Michigan show that it is possible to create an area of main memory that is as safe as disk from operating system crashes. This paper explores how to integrate the reliable memory provided by the Rio file cache into a database system. Prior studies have analyzed the performance benefits of reliable memory; we focus instead on how different designs affect reliability. We propose three designs for integrating reliable memory into databases: non-persistent database buffer cache, persistent database buffer cache, and persistent database buffer cache with protection. Non-persistent buffer caches use an I/O interface to reliable memory and require the fewest modifications to existing databases. However, they waste memory capacity and bandwidth due to double buffering. Persistent buffer caches use a memory interface to reliable memory by mapping it into the database address space. This places reliable memory under complete database control and eliminates double buffering, but it may expose the buffer cache to database errors. Our third design reduces this exposure by write protecting the buffer pages. Extensive fault tests show that mapping reliable memory into the database address space does not significantly hurt reliability. This is because wild stores rarely touch dirty, committed pages written by previous transactions. As a result, we believe that databases should use a memory interface to reliable memory.

Joint evaluation of recovery and performance of a COTS DBMS in the presence of operator faults

“…Very few works have evaluated experimentally the behavior of DBMS in the presence of faults. However, all the experimental evaluation works known in the literature have shown that a non-negligible number of faults are not handled correctly by DBMS [2][3][4][5][6]. While previous papers are "classic" fault injection works that have injected hardware and software faults, our work is, to the best of our knowledge, the first experimental evaluation of a DBMS with operator faults.…”

Cache Management of Big Data in Equipment Condition Assessment

“…One cluster is denoted as C and the data volume of C is denoted as C d . In the paper we propose a clustering algorithm called equal-capacity clustering algorithm based on k-means [12]. The algorithm divides all units of cache into the number of nodes (n), and each node has the same data volume.…”

Data Management in Non-Volatile Memory

“…Prior to NVM, researchers proposed battery-backed DRAM and buffer manager extensions to support recoverability [28]. For instance, [10] uses battery-backed DRAM with an ARIESlike protocol, and assumes page-level I/O for data and log updates.…”