This report describes simple hardware techniques for the measurement of the performance of multiprocessor computers. A number of examples of data obtained using these techniques are reported, as well as a discussion of the timing accuracy obtainable with this approach.

We have modified the C language to support a programming model based on a shared address space with physically distributed memory. With this model users can write programs in which the nodes of a massively parallel processor can access remote memory without message passing. AC pro vides support for distributed arrays as well as pointers to distributed data. Simple array references and pointer dereferencing are sufficient to generate low-overhead remote reads and writes.We have implemented these ideas in a compiler based on the GNU C compiler and targeted at Cray Research's T3D.Initial performance measurements show that AC generates code for remote accesses which is considerably faster than that of the native compiler for structures up to about 16 words in size and virtually equivalent for larger transfers.

No abstract

No abstract

Selecting a database management system involves matching users' requirements and the capabilities of available products. One way to simplify this task is to define data models identifying both data structures and the operations on those structures.In the past every commercial product has implemented its own data model. Now technical committee X3H2 of the American National Standards Institute is working on specifications for two models that are similar but not identical to many existing products.The network model is a structure-oriented model that is especially suitable for databases with static structures and a high volume of record-at-a-time processing.The relational model depends more heavily on operations than structures and thus provides the flexibility to handle dynamic databases.Examples written in the draft Network Database Language and the Relational Database Language demonstrate that both models can answer complex queries in a straightforward manner.In addition to the issue of data models, prospective buyers of database software need to consider features that affect daily operations. Existing hardware and operating systems sometimes limit the choice to a few commercial products. Systems also vary widely in their facilities for backup and recovery, bulk loading, schema manipulation, concurrency control, and report writers.