This paper describes a Distributed Shared Memory (DSM) system called Proteus, which aims to support runtime node reconfiguration. To make the system execute efficiently after node reconfiguration and reduce the overhead of reconfiguration, Proteus employs an adaptive workload redistribution strategy to perform workload redistribution as well as an affinity page movement strategy to redistribute memory pages. Furthermore, a forced update technique and a set of scalable synchronization mechanisms are developed to keep memory consistent and to perform a correct synchronization, respectively.

Exhaustive computation by a computer was used to find the unequal error protection capabilities of all binary cyclic codes of odd length up to 65 that have minimum distances of at least 3. For those codes for which upper bounds can only be computed on their unequal error protection capabilities, an analytic method developed by V.N. Dynkin and V.A. Togonidze (1976) is used to show that the upper bounds meet the exact unequal error protection capabilities.>