This paper presents the integration of two collective I/O techniques into the Clusterfile parallel file system: disk-directed I/O and two-phase I/O. We show that global cooperative cache management improves the collective I/O performance. The solution focuses on integrating disk parallelism with other types of parallelism: memory (by buffering and caching on several nodes), network (by parallel I/O scheduling strategies) and processors (by redistributing the I/O related computation over several nodes). The performance results show considerable throughput increases over ROMIO's extended two-phase I/O.

Finding changed identifiers in programs is important for program comparison and merging. Comparing two versions of a program is complicated if renaming has occurred. Textual merging is highly unreliable if, in one version, identifiers were renamed, while in the other version, code using the old identifiers was added or modified. A tool that automatically detects renamed identifiers between pairs of program modules is presented. The detector is part of a suite of intelligent differencing and merging programs that exploit the static semantics of programming languages. No special editor is needed for tracking changes. The core of the renaming detector is language independent. The detector works with multiple file pairs, taking into account renamings that affect multiple files. Renaming detectors for Java and Scheme have been implemented. A case study is presented that demonstrates proof of concept. With renaming detection, a higher quality of program comparison and merging is achievable.