Recovering disk storage metrics from low‐level trace events

Block devices such as magnetic disks are nonvolatile data storage devices that transfer data in fixed‐size chunks. They are the main nonvolatile memory that holds the file system, and they are also used in virtual memory mechanisms such swapping and page fault handling. Investigating storage performance issues requires a full insight into the operating system internals. Kernel tracing offers an efficient mechanism to gather information about the storage subsystem at runtime. Still, the tracing output is often huge and difficult to analyze manually.

[1]  Brendan Gregg.  Visualizing System Latency , 2010, ACM Queue.

[2]  Hewlett-Packard,et al.  Workload Dependent Performance Evaluation of the Linux 2 . 6 I / O Schedulers , 2004 .

[3]  Michel Dagenais,et al.  Operating System Level Trace Analysis for Automated Problem Identification , 2011 .

[4]  Margo I. Seltzer,et al.  Disk Scheduling Revisited , 1990 .

[5]  M. Desnoyers.  Low-Impact Operating System Tracing , 2009 .

[6]  Michel Dagenais,et al.  A Stateful Approach to Generate Synthetic Events from Kernel Traces , 2012, Adv. Softw. Eng..

[7]  Brendan Gregg,et al.  Dtrace: Dynamic Tracing in Oracle Solaris, Mac OS X and Freebsd , 2011 .

[8]  D. Toupin.  Using Tracing to Diagnose or Monitor Systems , 2011, IEEE Software.

[9]  Béchir Ktari,et al.  A complete framework for kernel trace analysis , 2011, 2011 24th Canadian Conference on Electrical and Computer Engineering(CCECE).

[10]  R. Krishnakumar.  Kernel korner: kprobes-a kernel debugger , 2005 .

[11]  Gregory R. Ganger,et al.  Towards higher disk head utilization: extracting free bandwidth from busy disk drives , 2000, OSDI.

[12]  Michel Dagenais,et al.  Lockless multi-core high-throughput buffering scheme for kernel tracing , 2012, OPSR.

[13]  Michel Dagenais,et al.  Efficient Model to Query and Visualize the System States Extracted from Trace Data , 2013, RV.

[14]  Erez Zadok,et al.  Benchmarking File System Benchmarking: It *IS* Rocket Science , 2011, HotOS.

[15]  Sameer Shende.  TAU Performance System® , 2011, Encyclopedia of Parallel Computing.

[16]  Robert Love,et al.  Linux Kernel Development , 2003 .

[17]  Yale N. Patt,et al.  Scheduling algorithms for modern disk drives , 1994, SIGMETRICS 1994.

[18]  Nikolai Joukov,et al.  A nine year study of file system and storage benchmarking , 2008, TOS.

[19]  Michel Dagenais,et al.  State History Tree: An Incremental Disk-Based Data Structure for Very Large Interval Data , 2013, 2013 International Conference on Social Computing.

[20]  Peter Druschel,et al.  Anticipatory scheduling: a disk scheduling framework to overcome deceptive idleness in synchronous I/O , 2001, SOSP.

[21]  Brendan Gregg,et al.  Systems Performance: Enterprise and the Cloud , 2013 .