Self-similarity in file systems

Steven D. Gribble, Gurmeet Singh Manku, Drew Roselli, Eric A. Brewer, Timothy J. Gibson, Ethan L. Miller
1998 Performance Evaluation Review  
We demonstrate that high-level le system events exhibit selfsimilar behaviour, but only for short-term time scales of approximately under a day. W e do so through the analysis of four sets of traces that span time scales of milliseconds through months, and that di er in the trace collection method, the lesystems being traced, and the chronological times of the tracing. Two s e t s o f detailed, short-term le system trace data are analyzed; both are shown to have self-similar like behaviour,
more » ... like behaviour, with consistent Hurst parameters a measure of self-similarity for all le system trafc as well as individual classes of le system events. Long-term le system trace data is then analyzed, and we discover that the traces' high variability and self-similar behaviour does not persist across time scales of days, weeks, and months. Using the short-term trace data, we s h o w that sources of le system trafc exhibit ON OFF source behaviour, which i s c haracterized by highly variably lengthed bursts of activity, followed by similarly variably lengthed periods of inactivity. This ON OFF behaviour is used to motivate a simple technique for synthesizing a stream of events that exhibit the same self-similar short-term behaviour as was observed in the le system traces.
doi:10.1145/277858.277894 fatcat:fauqmy5j4bfiva3ckzrn22zpli