Search
Join ACM

Computing Profession

Advertisement

Research and Advances

Proving the correctness of heuristically optimized code

A system for proving that programs written in a high level language are correctly translated to a low level language is described. A primary use of the system is as a postoptimization step in code generation. The low level language programs need not be generated by a compiler and in fact could be hand coded. Examples of the usefulness of such a system are given. Some interesting results are the ability to handle programs that implement recursion by bypassing the start of the program, and the detection and pinpointing of a wide class of errors in the low level language programs. The examples demonstrate that optimization of the genre of this paper can result in substantially faster operation and the saving of memory in terms of program and stack sizes.
Research and Advances

Shallow binding in Lisp 1.5

Shallow binding is a scheme which allows the value of a variable to be accessed in a bounded amount of computation. An elegant model for shallow binding in Lisp 1.5 is presented in which context-switching is an environment tree transformation called rerooting. Rerooting is completely general and reversible, and is optional in the sense that a Lisp 1.5 interpreter will operate correctly whether or not rerooting is invoked on every context change. Since rerooting leaves assoc [v, a] invariant, for all variables v and all environments a, the programmer can have access to a rerooting primitive, shallow[], which gives him dynamic control over whether accesses are shallow or deep, and which affects only the speed of execution of a program, not its semantics. In addition, multiple processes can be active in the same environment structure, so long as rerooting is an indivisible operation. Finally, the concept of rerooting is shown to combine the concept of shallow binding in Lisp with Dijkstra's display for Algol and hence is a general model for shallow binding.
Research and Advances

Automatic error recovery for LR parsers

In this paper we present a scheme for detecting and recovering from syntax errors in programs. The scheme, which is based on LR parsing, is driven by information which is directly and automatically obtainable from the information that is already present in an LR parser. The approach, which is patterned after that of Levy and Graham and Rhodes, appears to provide error recovery which is both simple and powerful.
and
Research and Advances

General equations for idealized CPU-I/O overlap configurations

General equations are derived for estimating the maximum possible utilization of main storage partitions, CPU and I/O devices under different conditions in an idealized CPU-I/O overlap model of multiprogrammed computer systems. The equations are directly applicable to any configuration consisting of sets of identical CPU's, I/O processors, main storage partitions and user tasks. Examples are provided to illustrate the use of the equations to compute effective processing time per record and expected timesharing response time under both balanced and unbalanced resource utilization conditions.
Research and Advances

Characteristics of application software maintenance

Maintenance and enhancement of application software consume a major portion of the total life cycle cost of a system. Rough estimates of the total systems and programming resources consumed range as high as 75-80 percent in each category. However, the area has been given little attention in the literature. To analyze the problems in this area a questionnaire was developed and pretested. It was then submitted to 120 organizations. Respondents totaled 69. Responses were analyzed with the SPSS statistical package. The results of the analysis indicate that: (1) maintenance and enhancement do consume much of the total resources of systems and programming groups; (2) maintenance and enhancement tend to be viewed by management as at least somewhat more important than new application software development; (3) in maintenance and enhancement, problems of a management orientation tend to be more significant than those of a technical orientation; and (4) user demands for enhancements and extension constitute the most important management problem area.
, , and
Research and Advances

Analyses of deterministic parsing algorithms

This paper describes an approach for determining the minimum, maximum, and average times to parse sentences acceptable by a deterministic parser. These quantities are presented in the form of symbolic formulas, called time-formulas. The variables in these formulas represent not only the length of the input string but also the time to perform elementary operations such as pushing, popping, subscripting, iterating, etc. By binding to the variables actual numerical values corresponding to a given compiler-machine configuration, one can determine the execution time for that configuration. Time-formulas are derived by examining the grammar rules and the program representing the algorithm one wishes to analyze. The approach is described by using a specific grammar that defines simple arithmetic expressions. Two deterministic parsers are analyzed: a top-down recursive descent LL(1) parser, and a bottom-up SLR(1) parser. The paper provides estimates for the relative efficiencies of the two parsers. The estimates applicable to a specific machine, the PDP-10, are presented and substantiated by benchmarks. Finally, the paper illustrates the proposed approach by applying it to the analyses of parsers for a simple programming language.
and
Research and Advances

Some basic determinants of computer programming productivity

The purpose of this research was to examine the relationship between processing characteristics of programs and experience characteristics of programmers and program development time. The ultimate objective was to develop a technique for predicting the amount of time necessary to create a computer program. The fifteen program characteristics hypothesized as being associated with an increase in programming time required are objectively measurable from preprogramming specifications. The five programmer characteristics are experience-related and are also measurable before a programming task is begun. Nine program characteristics emerged as major influences on program development time, each associated with increased program development time. All five programmer characteristics were found to be related to reduced program development time. A multiple regression equation which contained one programmer characteristic and four program characteristics gave evidence of good predictive power for forecasting program development time.
Research and Advances

Automated welfare client-tracking and service integration: the political economy of computing

The impacts of an automated client-tracking system on the clients, caseworkers, administrators, and operations of the welfare agencies that use it are reported. The major impact of this system was to enhance the administrative attractiveness of the using agencies in the eyes of funders rather than to increase their internal administrative efficiency. This impact is a joint product of both the technical features of the computer-based system and of the organizational demands placed upon different agencies, administrators, and caseworkers. It illustrates the way “successful” automated information systems fit the political economies of the groups that use them.
Research and Advances

Test data as an aid in proving program correctness

Proofs of program correctness tend to be long and tedious, whereas testing, though useful in detecting errors, usually does not guarantee correctness. This paper introduces a technique whereby test data can be used in proving program correctness. In addition to simplifying the process of providing correctness, this method simplifies the process of providing accurate specification for a program. The applicability of this technique to procedures and recursive programs is demonstrated.
Research and Advances

Automatic data structure selection: an example and overview

The use of several levels of abstraction has proved to be very helpful in constructing and maintaining programs. When programs are designed with abstract data types such as sets and lists, programmer time can be saved by automating the process of filling in low-level implementation details. In the past, programming systems have provided only a single general purpose implementation for an abstract type. Thus the programs produced using abstract types were often inefficient in space or time. In this paper a system for automatically choosing efficient implementations for abstract types from a library of implementations is discussed. This process is discussed in detail for an example program. General issues in data structure selection are also reviewed.
Research and Advances

A fast algorithm for copying list structures

An algorithm is presented for copying an arbitrarily linked list structure into a block of contiguous storage locations without destroying the original list. Apart from a fixed number of program variables, no auxillary storage, such as a stack, is used. The algorithm needs no mark bits and operates in linear time. It is shown to be significantly faster than Fisher's algorithm, the fastest previous linear-time algorithm for the same problem. Its speed comes mainly from its efficient list-traversal technique, which folds the processing stack into the structure being built, and from its classification of list cells into nine types, which enables processing operations to be optimized for each type.
Research and Advances

The SL5 procedure mechanism

This paper describes an integrated procedure mechanism that permits procedures to be used as recursive functions or as coroutines. This integration is accomplished by treating procedures and their activation records (called environments) as data objects and by decomposing procedure invocation into three separate components at the source-language level. In addition, argument binding is under the control of the programmer, permitting the definition of various methods of argument transmission in the source language itself. The resulting procedure mechanism, which is part of the SL5 programming language, is well suited to goal-oriented problems and to other problems that are more readily programmed by using coroutines. Several examples are given.
and
Research and Advances

A language extension for expressing constraints on data access

Controlled sharing of information is needed and desirable for many applications and is supported in operating systems by access control mechanisms. This paper shows how to extend programming languages to provide controlled sharing. The extension permits expression of access constraints on shared data. Access constraints can apply both to simple objects, and to objects that are components of larger objects, such as bank account records in a bank's data base. The constraints are stated declaratively, and can be enforced by static checking similar to type checking. The approach can be used to extend any strongly-typed language, but is particularly suitable for extending languages that support the notion of abstract data types.
and
Research and Advances

Optimal conversion of extended-entry decision tables with general cost criteria

A general dynamic programming algorithm for converting limited, extended, or mixed entry decision tables to optimal decision trees is presented which can take into account rule frequencies or probabilities, minimum time and/or space cost criteria, common action sets, compressed rules and ELSE rules, sequencing constraints on condition tests, excludable combinations of conditions, certain ambiguities, and interrupted rule masking.
Research and Advances

List processing in real time on a serial computer

A real-time list processing system is one in which the time required by the elementary list operations (e.g. CONS, CAR, CDR, RPLACA, RPLACD, EQ, and ATOM in LISP) is bounded by a (small) constant. Classical implementations of list processing systems lack this property because allocating a list cell from the heap may cause a garbage collection, which process requires time proportional to the heap size to finish. A real-time list processing system is presented which continuously reclaims garbage, including directed cycles, while linearizing and compacting the accessible cells into contiguous locations to avoid fragmenting the free storage pool. The program is small and requires no time-sharing interrupts, making it suitable for microcode. Finally, the system requires the same average time, and not more than twice the space, of a classical implementation, and those space requirements can be reduced to approximately classical proportions by compact list representation. Arrays of different sizes, a program stack, and hash linking are simple extensions to our system, and reference counting is found to be inferior for many applications.

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More