Program-Counter-Based Pattern Classification in Buffer Caching

Program-counter-based (PC-based) prediction techniques have been shown to be highly effective and are widely used in computer architecture design. In this paper, we explore the opportunity and viability of applying PC-based prediction to operating systems design, in particular, to optimize buffer caching. We propose a Program-Counterbased Classification (PCC) technique for use in patternbased buffer caching that allows the operating system to correlate the I/O operations with the program context in which they are issued via the program counters of the call instructions that trigger the I/O requests. This correlation allows the operating system to classify I/O access pattern on a per-PC basis which achieves significantly better accuracy than previous per-file or per-application classification techniques. PCC also performs classification more quickly as per-PC pattern just needs to be learned once. We evaluate PCC via trace-driven simulations and an implementation in Linux, and compare ...

• Type: Conference
• Year: 2004
• Where: OSDI
• Authors: Chris Gniady, Ali Raza Butt, Y. Charlie Hu
• Keywords: Buffer Caching, Operating System, OSDI 2004, Per-application Classification Techniques, State-ofthe-art Pattern-based Buffer