Brian T. Bennett

LRU stack processing

Stack processing, and in particular stack processing for the least recently used replacement algorithms, may present computational problems when it is applied to a sequence of page references with many different pages. This paper describes a new technique for LRU stack processing that permits efficient processing of these sequences. An analysis of the algorithm and a comparison of its running times with those of the conventional stack processing algorithms are presented. Finally we discuss a multipass implementation, which was found necessary to process trace data from a large data base system.

SCHOOL BUS ROUTING BY COMPUTER

A PROCEDURE IS DESCRIBED FOR THE DESIGNING OF SCHOOL BUS ROUTES BY COMPUTER. THE PROCEDURE IS MORE FLEXIBLE THAN PREVIOUSLY PUBLISHED ONES IN A NUMBER OF WAYS. THE OBJECTIVE FUNCTION INCLUDES THE REDUCTION OF STUDENT TRAVEL TIME, U-TURNS, AND BUS OVERLOADING, AS WELL AS THE REDUCTION OF BUS TRAVEL TIME. THE PROCEDURE CAN ALSO HANDLE ONE-WAY STREETS, A BUS GARAGE OTHER THAN AT THE SCHOOL SITE AND THE REQUIREMENT THAT AT CERTAIN STOPS THE BUS MUST PICK UP CHILDREN FROM ONE PARTICULAR SIDE OF THE ROAD. THE PROCEDURE IS AN EXTENSION OF THE CLARKE AND WRIGHT ALGORITHM FOR SCHEDULING DELIVERY VEHICLES. /TRRL/

Adaptive variation of the transfer unit in a storage hierarchy

Consider a paged storage hierarchy with at least two levels L1 and L2 where L1 denotes main storage and L2 secondary storage. Suppose that the unit of replacement for L1 is a single page, and that the L2-to-L1 transfer unit, given a page fault, is an integer number of pages. Then, given a suitable replacement policy for L1 increasing the unit of transfer often results in a lower miss ratio at the expense of increased paging traffic. This paper explores the possibility of adaptively varying the L2-to-L1 transfer unit as a function of the reference history of the data to be fetched. Experiments on traces drawn from two large data base systems suggest that such adaptation can result in improved tradeoffs between miss ratios and paging traffic.

Permutation clustering: an approach to on-line storage reorganization

A class of dynamic reorganization algorithms is described which embodies a number of desirable systems properties. Experiments on a trace taken from a large data base application indicate that a member of this class may be used to obtain time-varying or quasistatic organizations that exhibit improved paging performance.

LINCOLN TUNNEL TRAFFIC CONTROL EXPERIMENT

AN EXPERIMENT TO DEVELOP COMPUTER METHODOLOGY FOR IMPROVING TRAFFIC FLOW THROUGH LINCOLN TUNNEL IS DESCRIBED. SOME REMARKS ARE GIVEN ON THE POTENTIAL SPINOFFS FROM THIS EXPERIMENT WHICH MAY INFLUENCE THE DEVELOPMENT OF OTHER TRAFFIC CONTROL SYSTEMS. RESULTS SHOWED A MEASURABLE INCREASE IN THROUGHPUT, HIGHER SPEEDS WITH LOWER DENSITIES DURING THE CONTROL PERIOD. IT WAS CONCLUDED THAT HIGH FLOWS COULD BE MAINTAINED AT MODERATELY HIGH SPEED IF DENSITIES DID NOT REACH CONGESTION LEVELS. THE EXPERIMENT HAD BENEFITS IN PROVIDING ACCURATE TRAFFIC DATA, CALIBRATION AND CORRECTION OF MAGNETIC LOOP DETECTORS. /HSL/