Guest Editors' Editorial

Editorial Commentary

Guest Editors' Editorial     

A report on the sisal language project

Sisal (Streams and Iterations in Single Assignment Language) is a general-purpose applicative language intended for use on both conventional and novel multiprocessor systems. In this report we discuss the project's objectives, philosophy, and accomplishments and state our future plans. Four significant results of the Sisal project are compilation techniques for high-performance parallel applicative computation, a microtasking environment that supports dataflow on conventional shared-memory architectures, execution times comparable to those of Fortran, and cost-effective speedup on shared-memory multiprocessors.     

Applicative parallelism on a shared-memory multiprocessor

An applicative language, SISAL (streams and iteration in a single-assignment language), is used to examine how effective the applicative approach can be in programming parallel computers. SISAL is a research language for investigating issues in parallel processing, especially for numerical computing. A prototype implementation of SISAL is described, focusing on loops and streams and on microtasking software. Performance results are reported for Livermore kernel 1, the Sieve prime finder, insertion sort, Batcher sort, the Simple hydrodynamics code, and the PSPhot photon-transport code     

Dataflow Resource Managers and Their Synthesis from Open Path Expressions

The control of concurrent access to shared resources is an important feature of both centralized and distributed operating systems. In conventional systems, exclusive access is the rule while concurrent access is the exception. Dataflow computer systems, along with an applicative style of programming, provide an execution environment in which this philosophy is reversed. In these latter systems, it is necessary to reexamine the manner in which synchronization of access to shared resources is specified and implemented. A basic design for a dataflow resource manager is reviewed, illustrating the clear separation between access mechanism and scheduling policy. The semantics of the access mechanism is based solely on the principle of data dependency. Specifications are presented for a general scheduler to further constrain or order accesses to the resource. Using ``open path expressions' as a very high-level specification language for synchronization, it is shown how to automatically synthesize a scheduler as a distributed network of communicating modules.     

A software scheme for user-controlled file encryption

An implementation of a technique, based on the Data Encryption Standard (DES), is described for user-controlled encryption of files. The user supplies a privately selected character string as a key and also the desired number of DES iterations. The actual encryption key and initial chaining value for block-chaining are derived from this private key for encryption of the user file. The implementation is adaptible for use on both time-sharing systems and personal computer systems.     

An Interactive Manager for Pascal Software

An interactive facility for development and maintenance of Pascal programs is described. Features include interactive addition, deletion and modification of modules, independentcompilation of groups of procedures, and automatic recompilation of an entire program only if global declarations are modified. The implementation required modification of a Pascal compiler to process an arbitrary number of programs in a single run and to compile correctly if the main program is missing. Overall processing is managed by a parameterized control statement procedure that invokes several system utilities. The facility has been used successfully in instructional and research computing situations. Significant remaining problems are errors caused by mismatched parameters between independently compiled procedures, and interpretation by users of error messages generated by system utilities.     

Success and failure analysis of a problem solving system implementation

The NAPSS (Numerical Analysis Problem Solving System) project was an attempt to develop an interactive system, with a high-level language resembling conventional mathematics, for solving problems in numerical analysis. This report uses NAPSS as an example to discuss some aspects of the design and implementation of the programming languages and software which comprise problem solving systems. Successful features as well as failures (with alternative proposals) are detailed in the areas of general language management, specific language features and the operating system interface. In general, a massive, coherent implementation of theoretical work being done at the frontiers of research has abundant opportunity for failure.     

The use of combinators in translating a purely functional language to low-level data-flow graphs

FCL is a higher-order functional programming language which consolidates and extends a number of desirable features of existing languages. This paper describes the salient features of FCL and an algorithm for translation to highly parallel data flow graphs. The translation algorithm is based on a set of extended “combinators”. The relationship between functional programming languages and demand-driven or data-driven data flow architectures is established.     

An analysis of program execution on a recursive stream-oriented data flow architecture

The functional or applicative approach to the development of software leads to the natural exposure of parallelism. Data flow architectures are designed to exploit this parallelism. This paper describes a method for modeling and analyzing program execution on a stream-oriented, recursive data flow architecture. Based on a maximum path length technique, a recursive analysis of nodes (corresponding to high level constructs) results in an approximation of the time required to execute the program given adequate resources. By relating the structure of a program to the degree of exhibited parallelism, this method provides a useful tool for studying parallelism and evaluating certain problem classes to be run on similar architecutures.