Making Early Predictions of File Accesses
نویسندگان
چکیده
Nearly all extant file access predictors attempt to identify the immediate successor to the file being currently accessed. As a result, they leave almost no time to prefetch the predicted file before it is accessed. We present here a Perceptron-based file predictor that identifies the files that will be accessed up to five accesses ahead. Experimental evidence shows that our early predictor can make between 30 and 78 percent of correct predictions, depending on the nature of the file system workload and the position of the access it attempts to predict.
منابع مشابه
Improving Data Availability Using Combined Replication Strategy in Cloud Environment
As grow as the data-intensive applications in cloud computing day after day, data popularity in this environment becomes critical and important. Hence to improve data availability and efficient accesses to popular data, replication algorithms are now widely used in distributed systems. However, most of them only replicate the static number of replicas on some requested chosen sites and it is ob...
متن کاملUsing program and user information to improve file prediction performance
Correct prediction of file accesses can improve system performance by mitigating the relative speed difference between CPU and disks. This paper discusses Program-based Last Successor (PLS) and presents Programand Userbased Last Successor (PULS), file prediction algorithms that utilize information about the program and user that access the files. Our simulation results show that PLS makes 21% f...
متن کاملQuantifying Temporal and Spatial Localities Due to File System Caching
Modern storage consists of many optimizations along the data path, from the front-end file system cache and disk layout management to the back-end disk controller cache and on-disk caching. Various components generally exploit the temporal and spatial localities of a workload to improve performance. However, how localities of a workload are transformed by each optimization is not well understoo...
متن کاملConflict Detection Algorithm to Minimize Locking for MPI-IO Atomicity
Many scientific applications require high-performance concurrent I/O accesses to a file by multiple processes. Those applications rely indirectly on atomic I/O capabilities in order to perform updates to structured datasets, such as those stored in HDF5 format files. Current support for atomicity in MPI-IO is provided by locking around the operations, imposing lock overhead in all situations, e...
متن کاملAnomaly Detection in Computer Security and an Application to File System Accesses
We present an overview of anomaly detection used in computer security, and provide a detailed example of a host-based Intrusion Detection System that monitors file systems to detect abnormal accesses. The File Wrapper Anomaly Detector (FWRAP) has two parts, a sensor that audits file systems, and an unsupervised machine learning system that computes normal models of those accesses. FWRAP employs...
متن کامل