Achieving high performance for parallel or distributed programs often requires substantial amounts of information about the programs themselves, about the systems on which they are executing, and about specific program runs. The monitoring system that collects, analyzes, and makes application-dependent monitoring information available to the programmer and to the executing program is presented. The system may be used for off-line program analysis, for on-line debugging, and for making on-line, dynamic changes to parallel or distributed programs to enhance their performance. The authors use a high-level, uniform data model for the representation of program information and monitoring data. They show how this model may be used for the specification of program views and attributes for monitoring, and demonstrate how such specifications can be translated into efficient, program-specific monitoring code that uses alternative mechanisms for the distributed analysis and collection to be performed for the specified views. The model's utility has been demonstrated on a wide variety of parallel machines.>

Abstract The installable file system (IFS) faculty on OS/2 offers an efficient mechanism for expanding the abilities of the OS/2 kernel. Two prototypes illustrate the usefulness of the IFS facility. An OS/2 port of the Andrew File System (AFS) allows users to access global community files while retaining the private local file system on the OS/2 machine. An IFS implementation of a network protocol mapper, which maps TCP/IP socket calls to SNA Advanced Program‐to‐Program Communication (APPC) calls, provides the ability to run any socket program on an APPC network after simply re‐linking with the mapping library. The only impediment to the development of these prototypes is the lack of kernel‐level interfaces to operating system devices, resulting in time‐consuming data movement back up to the user level to access the network and the local disks. This impediment will disappear as OS/2 and its associated support software matures.

The programming of efficient parallel software typically requires extensive experimentation with program prototypes. To facilitate such experimentation, any programming system that supports rapid prototyping of parallel programs should provide high-level language primitives with which programs can be explicitly, statically, or dynamically tuned with respect to performance and reliability. Such language primitives should be able to refer conveniently to the information about the executing program and the parallel hardware required for tuning. Such information may include monitoring data about the current or previous program or even hints regarding appropriate tuning decisions. Language primitives and an associated programming system for program tuning are presented. The primitives and system have been implemented, and have been tested with several parallel applications on a network of Unix workstations.>