A Scalable Distributed Multimedia Knowledge Retrieval System on a Cluster of Heterogeneous High Performance Architectures
نویسندگان
چکیده
instructions Control stack Program Data parallel environment communication association Fig. 2. A heterogeneous associative model. represented as an associative frame, and uses associative vectors to facilitate fast backtracking. The vectors are indexed with respect to the base of the current frame. Variable aliases are indicated by filter vectors and are tracked by the alias management table. The logical OR of two vectors creates the union of two sets of aliased variables when members of the two sets are aliased by an instruction. A detailed explanation of this model and the corresponding abstract instructions is given in [4]. A shared variable has two occurrences in a clause body: a producer occurrence and a consumer occurrence. The producer occurrence generates a value for the shared variable, and the consumer occurrence uses the value of a variable. 2.2. Notations for the abstract instructions In this section, we define the notations used in the examples. We will show examples of the abstract instructions code for both the abstract machine of the server processes and the distributed abstract machine used by the coordinator. A symbol denotes the start of a procedure with an arity of . A symbol _CN denotes the start of the Nth clause of the predicate . A symbol denotes a universal filter vector a vector of Boolean values associated with a set of clauses to identify unifiable clauses. A symbol denotes a binding filter vector a vector of Boolean values marking the values in the clause-head representation to which a variable in a goal will be bound after unification. A symbol denotes a fact vector a vector of Boolean values marking the facts in a knowledge base. A symbol denotes a clause filter vector a vector of Boolean values marking clauses with non-empty body. A symbol denotes the Nth argument of the current goal. A symbol denotes a temporary Boolean Filter vector. A symbol indicates a global register, and a symbol denotes a variable. A symbol indicates a Boolean filter vector marking the servers which contain a particular predicate, and a symbol denotes a label. Example 1 illustrates a simple logic program. The corresponding compiled abstract instruction code is given in Figure Stephen W. Ryan and Arvind K. Bansal
منابع مشابه
Applying Java for the Retrieval of Multimedia Knowledge Distributed on High Performance Clusters on the Internet
This paper describes a Java application for interfacing with an Internet based distributed multimedia knowledge retrieval system. The aim of this system is to facilitate the access of distributed knowledge residing on remote computer systems. The motivation for this work is the increasing demand for applications that take advantage of the Internet and local intranets for accessing and integrati...
متن کاملA Scalable Distributed Associative Multimedia Knowledge Base System for the Internet
This paper describes a system to retrieve multimedia knowledge on a cluster of heterogeneous architectures distributed over the internet. In this system, knowledge is represented using facts and rules in an associative logic-programming model. Associative computing facilitates search by content and exploits data parallel computation. It also provides a flat data model that can be easily mapped ...
متن کاملDesign of a novel congestion-aware communication mechanism for wireless NoC architecture in multicore systems
Hybrid Wireless Network-on-Chip (WNoC) architecture is emerged as a scalable communication structure to mitigate the deficits of traditional NOC architecture for the future Multi-core systems. The hybrid WNoC architecture provides energy efficient, high data rate and flexible communications for NoC architectures. In these architectures, each wireless router is shared by a set of processing core...
متن کاملA Scalable Distributed Heterogeneous Associative Logic Programming System
This paper describes a distributed implementation of a scalable heterogeneous associative logic programming model, and describes an abstract instruction set for the distributed version of the model. Associative computation exploits data parallel computation. The implementation uses PVM for architecture independence, and uses object oriented programming for modularity and portability. Performanc...
متن کاملA MapReduce Based Distributed LSI for Scalable Information Retrieval
Latent Semantic Indexing (LSI) has been widely used in information retrieval due to its efficiency in solving the problems of polysemy and synonymy. However, LSI is notably a computationally intensive process because of the comput260 Y. Liu, M. Li, M. Khan, M. Qi ing complexities of singular value decomposition and filtering operations involved in the process. This paper presents MR-LSI, a MapR...
متن کاملTowards a Scalable Networked Retrieval System for Searching Multimedia Databases
In this paper the architecture of a distributed and scalable multimedia information retrieval system (Dsmily) is described. The system consists of hierarchically organized networked nodes and is designed to integrate existing dynamic multimedia databases. The document ranking process as well as the preselection of databases to be searched, both tasks are based on a probabilistic model for distr...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- International Journal on Artificial Intelligence Tools
دوره 9 شماره
صفحات -
تاریخ انتشار 2000