A light-weight, temporary file system for large-scale Web servers

Several recent studies have pointed out that file I/Os can be a major performance bottleneck for some large Web servers. Large I/O buffer caches often do not work effectively for large servers. This paper presents a novel, light-weight, Temporary File System called TFS that can effectively im-prove I/O performance for large servers. TFS is a more cost-effective scheme compared to the full caching policy for large servers. It is a user-level application that manages files on a raw disk or raw disk partition and works in con-junction with a file system as an I/O accelerator. Since the entire system works in the user space, it is easy and inexpen-sive to implement and maintain. It also has good portabil-ity. TFS uses a novel disk storage subsystem called Cluster-structured Storage System (CSS) to manage files. CSS uses only large disk reads and writes and does no have garbage collection problems. Comprehensive trace-driven simulation experiments show that, TFS achieves up to 160 % better sys-tem throughput and reduces up to 77 % I/O latency per URL operation than that in a traditional Unix Fast File System in large Web servers.