| 1. |
- Ahlgren, Bengt, et al.
(författare)
-
A Host Interface to the DTM Network
- 1992. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- DTM, dynamic synchronous transfer mode, is a new time division multiplexing technique for fiber networks currently being developed and implemented at the Royal Institute of Technology in Stockholm, Sweden. This paper describes the hardware and software aspects of the design of an SBus host interface to the DTM network for a Sun SPARCstation. The interface is based on a dual port memory residing on the interface card and accesible over the SBus from the host CPU. The host operating system allocates message buffers directly in this memory. The interface has hardware support for segmenting and reassembling packets to and from the data units of the DTM. The software part of the interface manages the shared memory and the virtual circuits provided by the DTM network.
|
|
| 2. |
- Ali, Khayri Mohammed
(författare)
-
A method for implementing out in parallel execution of Prolog
- 1987. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- A method for implementing cut in parallel execution of Prolog is presented. It takes advantages of the efficient implementation of cut in the sequential WAM. It restricts the parallelism, however, it is simple and adds a small extra overhead over the sequential scheme. The method can be used in parallel execution of Prolog on shared amd nonshared memory multiprocessors.
|
|
| 3. |
- Ali, Khayri Mohammed, et al.
(författare)
-
An investigation of an OR parallel execution model for horn clause programs
- 1988. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- We present a model for OR parallel execution of Horn clause programs on a combined local and shared memory multiprocessor system. In this model, the shared memory only contains control information that guides processors requesting a job to independently construct the environment required to get a new job. Each processor has a local memory containing its own binding environment. This reduces the traffic to the shared memory and allows each processor to process its job with high performance. Each processor is almost the same as Warren's Abstract Machine (WAM). A method for nonshared memory multiprocessor architectures is outlined. We also present some preliminary results of an experimental investigation of the model.
|
|
| 4. |
- Ali, Khayri Mohammed, et al.
(författare)
-
Global garbage collection for distributed heap storage systems
- 1987. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- We present a garbage-collection algorithm, suitable for loosely-coupled multiprocessor systems, in which the processing elements (PE's) share only the communication medium. The algorithm is global, i.e. it involves all the PE's in the system. It allows space compaction, and it uses a system-wide marking phase to mark all accessible objects where a combination of parallel breadth-first/depth-first strategies is used for tracing the object-graphs according to a decentralized credit mechanism that regulates the number of garbage collection messages in the system. The credit mechanism is crucial for determining the space requirement of the garbage-collection messages. Also a variation of the above algorithm is presented for systems with high locality of reference. It allows each PE to perform first its local garbage collection and only invokes the global garbage collection when the freed space by the local collector is insufficient.
|
|
| 5. |
- Ali, Khayri Mohammed
(författare)
-
OR parallel execution of horn clause programs based on WAM and shared control information
- 1988. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- A method for OR parallel execution of Horn clause programs on a shared memory multiprocessor is presented. The shared memory contains only control information that guide processors requesting a job to independently construct the environment required to get a new job from the other processors without degrading the performance. Each processor has a local memory containing its own binding environment. This reduces the traffic to the shared memory and allows each processor to process its job with high performance. Each processor is almost the same as the Warren Abstract Machine (WAM). Modification to the WAM for supporting the method is described in detail. A method for nonshared memory multiprocessor architecture is outlined.
|
|
| 6. |
- Ali, Khayri Mohammed
(författare)
-
OR-Parallel Execution of Prolog on a Multi-Sequential Machine
- 1986. - 1
-
Rapport (refereegranskat)abstract
- Based on extending the sequential execution model of Prolog to include parallel execution, we present a method for OR-parallel execution of Prolog on a multiprocessor system. The method reduces the overhead incurred by parallel processing. It allows many processing elements (PEs) to process simultaneously a common branch of a search tree, and each of these PEs creates its local environment and selects a subtree for processing without communication. The run-time overhead is small: simple and efficient operations for selecting the proper subtree. Communication is necessary only when some PEs have exhausted their search spaces and there are others still searching for solutions. The method is able to utilize most of the technology devised for sequential implementation of Prolog. It is optimized for an architecture which supports broadcast copying.
|
|
| 7. |
- Ali, Khayri Mohammed
(författare)
-
OR parallel execution of Prolog on BC-machine
- 1988. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- We present the principles of OR-parallel execution of Prolog on a special parallel inference machine, named BC-machine. The machine is a combined local and shared memory multiprocessor with a special interconnection network. The network allows write operations of an active processor to be broadcasted to several idle processors simultaneously. The shared memory is mainly used for sharing some control information among processors in the system. The amount of shared control information is small and accessed relatively seldom. The execution model is based on the local address space approach. It allows all the technology developed for standard Prolog to be used without loss of efficiency. We expect that the model substantially reduces the copying overhead in comparision with all previous related models. This reduction is due to our new idea of dynamic load balancing and the broadcast mechanism supported by the network. A prototype system of 9 processors is under construction at the Swedish Institute of Computer Science in Stockholm.
|
|
| 8. |
- Ali, Khayri Mohammed, et al.
(författare)
-
Performance of Muse on Switch-Based Multiprocesor Machines
- 1992. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- The Muse (multiple sequential Prolog engines) approach has been used to make a simple and efficient OR-parallel implementation of the full Prolog language. The performance results of the Muse system on bus-based multiprocessor machines have been presented in previous chapters, papers. This chapter paper discusses the implementation and performance results of the Muse system on switch-based multiprocessors (the BBN Butterfly GP1000 and TC2000). The results of Muse execution show that high real speedups can be achieved for Prolog programs that exhibit coarse-grained parallelism. The scheduling overhead is equivalent to around 8 -- 26 Prolog procedure calls per task on the TC2000. The chapter paper also compares the Muse results with corresponding results for the Aurora OR-parallel Prolog system. For a large set of benchmarks, the results are in favor of the Muse system.
|
|
| 9. |
- Ali, Khayri Mohammed, et al.
(författare)
-
The Engine-Scheduler Interface used in the Muse OR-parallel Prolog System
- 1992. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- Almost any sequential Prolog system is in principle easy to extend for OR-parallelism, using the Muse execution model. To reduce your programming effort we have implemented the Muse scheduler, with a clean interface to the Prolog sequential engine. This interface is implemented as a set of C macros. The sequential Prolog system to be parallelized uses some of those macros provided by the Muse scheduler and must also provide some macros for the Muse scheduler. This chapter paper contains a definition and description of the required macros, emphasizing information needed by the Prolog engine programmer.
|
|
| 10. |
- Alshawi, Hiyan, et al.
(författare)
-
Bilingual conversation interpreter : a prototype interactive message translator. Final report
- 1991. - 1
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- This document is the final report for a research project aimed at producing a prototype system for on-line translation of typed dialogues between speakers of different natural languages. The work was carried out jointly by SICS and SRI Cambridge. The resulting prototype system (called Billingual Conversation Interpreter, or BCI) translates between english and Swedish in both directions.The Major components of the BCI are two copies of the SRI Core Language Engine, equipped with English and Swedish grammars respectively. These are linked by the transfer and disambiguation components. Translation takes place by analyzing the source-language sentence into Quasi Logical Form ( QLF), a linguistically motivated logical representation, transferring this into a target-language QLF, and generating a target-language sentence. When ambiguities occur that cannot be resolved automatically, they are clarified by Querying the appropriate user. The clarification dialogue presupposes no knowledge of either linguistics or the other language. The prototype system has a broad grammatical coverage, a initial vocabulary of about 1000 words together with vocabulary expansion tools, and a set of English-Swedish transfer rules. The formalism developed for coding this linguistic information make it relatively easy to extend the system. We believe that the project was successful in demonstrating the feasibility of using these techniques for interactive translation applications, and provides a sound basis for development of a large scale message translator system with potential for commercial exploitation.The main sections of this report are the following: * A non-technical introduction, summarizing the BCI's design , and containing a sample session. * An overview of the Swedish version of the CLE. * A detailed discussion of the theory and practice of QLF transfer. * A description of the interactive disambiguation method. * Suggestions for possible follow-on projects aimed in the direction of practically usable commercial systems.
|
|
