| 1. |
- Ciepielewski, Andrzej, et al.
(författare)
-
Performance evaluation of a storage model for OR-parallel execution of logic programs
- 1986. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- As the next step towards a computer architecture for parallel execution of logic programs we have implemented four refinements of the basic storage model for OR-Parallelism and gathered data about their performance on two types of shared memory architectures, with and without local memories. The results show how the different properties of the implementations influence performance, and indicate that the implementations using hashing techniques (hash windows) will perform best, especially on systems with a global storage and caches. We rise the question of the usefulness of the simulation technique as a tool in developing new computer architectures. Our answer is that simulations can not give the ultimate answers to the design questions, but if only the judiciosly chosen parts of the machine are simulated on a detailed level, then the obtained results can give a very good guidance in making design choice.
|
|
| 2. |
- Hausman, Bogumil
(författare)
-
A Simulator of the OR-Parallel Token Machine
- 1986. - 1
-
Rapport (refereegranskat)abstract
- This report is mainly meant as the documentation of the simulator of the OR-Parallel Token Machine for Horn Clause programs. The simulator has been used to investigate the dynamic characteristics of pure Horn Clause programs and to evaluate several storage structures. We start by briefly describing the virtual machine. Then we discuss the merits of a specification language Meta IV and a programming language, Simula, and also show transformations between those two. Finally, we describe the constituent parts of the simulation system, namely the underlying message passing mechanism and the three components of the machine: instruction processor, token pool and storage. The mapping of the specification into Simula shows the power of using object oriented languages for implementing abstract specifications and for simulation purposes.
|
|
| 3. |
- Hausman, Bogumil, et al.
(författare)
-
OR-parallel Prolog made efficient on shared memory multiprocessors
- 1987. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- With the arrival of commercially available shared-memory multiprocessors, Prolog implementation efforts begin to shift from single processor architectures to the new ones. Among the main problems are efficient implementation of operations on variables and of task switching. Most of the solutions proposed so far suffer from expensive, non-constant time implementation of operations on variables. We propose a model (Versions-Vector Model) in which operations on all variables are constant time operations. The price we pay is a non-constant time of a task switch. As a remedy we propose two ways of decreasing that price. The first is promotion of variables on a task switch, from versions-vectors to the stack or heap, making subsequent task switches cheaper. The second is delayed installation of variables in versions-vectors, decreasing the cost of short branches. We believe that the increased memory consumption induced by our model can be accepted as it is traded for speed.
|
|
| 4. |
- Lusk, E., et al.
(författare)
-
The Aurora or-parallel Prolog system
- 1990. - 2
-
Ingår i: New generation computing. - : ICOT, Tokyo, Japan. - 0288-3635 .- 1882-7055. ; 7:2,3, s. 243-271
-
Tidskriftsartikel (refereegranskat)
|
|
