R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. In The VLDB Journal, 3:161-- 210, 1994.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....of the behaviour of the solver which again was made possible thanks to the specific programming style (in particular continuations and memoisation) being used. As an alternative to our approach we considered using o# the shelf implementations of deductive databases, e.g. the Coral system [26], or logic programming systems tuned to find all solutions, e.g. XSB Prolog [11] However we found this to be a less viable approach in order to sustain our overall objective of automatic complexity analysis [22] Acknowledgements. The experiments reported in Section 5 were carried out in ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161--210, 1994.

....systems such as Nail [26] LOLA [12] Glue Nail [10] XSB [33] Aditi [37] LogicBase [13] and Declare SDS [16] only support flat relations which are found inappropriate for advanced applications. A few deductive database systems that support data with complex structures such as LDL [8] and CORAL [30] are only implemented as memory based systems and do not support direct access of data with complex structures [23] Since 1996, we have designed and implemented a persistent deductive database system called Relationlog. Being a persistent deductive system, Relationlog naturally supports ....

....the meta information about relations and rules. Then it uses one of three kinds of evaluation strategies to process the user query: 1) matching; 2) semi naive bottom up evaluation; 3) top down evaluation. Matching Most deductive database systems such as Aditi [37] LOLA [12] LDL [8] CORAL [30], etc, use magic set rewriting techniques to improve the performance of queries that involves rules. Basically, they focus on how to improve the evaluation speed by using the constants provided in the queries. There are several problems with this approach. First, if there is no constant in the ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161--210, 1994.

....manner for better management and easy retrieval. However, there are a number of such database systems, such as object oriented database systems [3, 4, 10] object relational database system [11] deductive database systems [7] and deductive object oriented object relational database systems [2, 6, 5, 8]. The question is which is suitable for storing CAD data. Based on our analysis, we believe that persistent deductive object relational or object oriented database systems are ideal for supporting CAD as they not only provide direct support for the e ective storage and ecient access to large ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. The VLDB Journal, 3(2):161-210, 1994.

....for storing relations as well as for organising sets of waiting consumers. The eciency, simplicity and expressiveness of the logic made it our favourite choice over the transformational approach of McAllester [11] or o the shelf implementations of deductive databases as, e.g. the CORAL system [14]. The complexity analysis has bene tted from the pioneering ideas of McAllester [11] on the complexity of solving (classical) Horn clauses. Here, we generalised these techniques to a richer class of input formulas and adapted it to the speci c properties of our solver. In doing this, we were ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161-210, 1994.

....purposes, such as a typical Prolog program, it is common to only need one answer to the query. The author has been partially supported by the Spanish CICYT (project TIC 980445 C03 02 TREND) One consequence of this is that most deductive database systems (for instance, DATALOG [25] CORAL [20], ADITI [26] use bottom up evaluation methods instead of top down one. Bottom up approach allows us to use set at a time evaluation, i.e. it processes sets of goals, rather than proceeding one (sub) goal at a time, where operations like relational joins can be made for disk resident data ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. In VLDB Journal, 3(2):161-210, 1994.

....closure declaratively in polynomial space and time [4, 5] which makes them expressive enough while still practical for real database applications. However, most of the research on such kind of deductive databases stays at the theoretical level. A few implemented systems such as LDL [10] and CORAL [20] are just memory based. In the past few years, we have designed and implemented a persistent deductive database system called Relationlog [16] The Relationlog system is based on the deductive database language Relationlog [17] which is a typed extension to Datalog with tuples and sets. It ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161-210, 1994.

....data with complex structures have been proposed, such as LDL [6] LPS [12] COL [1] Hilog [5] Relationlog [16] See [17] for an overview of some of these languages. Also, several deductive database systems that support data with complex structures have been developed, such as LDL [6] CORAL [23], and Relationlog [19] Object oriented concepts have evolved in three di erent disciplines: rst in programming languages, then in arti cial intelligence, and then in databases since the end of the 60 s. Indeed, object orientation o ers some of the most promising ways to meet the demands of many ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. The VLDB Journal, 3(2):161-210, 1994.

....languages (e.g. Parlog) to DBMSs [2] and to compiled, set at a time processing of logic queries against a database. Some of the significant efforts that also involve significant prototyping and in some cases performance evaluation include BERMUDA [6] BrAID [17] the LDL system [1] and Coral [14, 18, 15]. Our choice of the LDL system was guided by the merits of the system in functional terms, especially the ability of the LDL system to access commercial DBMSs of interest to us and its interface to C , as well as very important pragmatic considerations, such as: availability of technical ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri "The CORAL Deductive System", The VLDB Journal, Volume 3(2), April 1994.

....object relational ones. Models are usually versioned, making techniques from temporal DBs of interest. Other architectures that combine OO and deductive capabilities in sophisticated ways can also provide significant benefits to model management such as, Telos [MBJK90] ConceptBase [JJ89] Coral [RSSS94] NAIL [MUG86] and F Logic [KLW95] Inferencing in model management: Several key operations in model management involve various forms of inference, such as inverting a mapping, completing a mapping, and determining equivalence of models. For example, a mapping can be thought of as a view of ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The CORAL deductive system. VLDB Journal, 3(2):161--210, 1994.

....perform relatively poorly on others. The ROL system automatically decides which mechanism to use without user s intervention, based on the nature of the query and its knowledge about the data in the object base. 4. 1 Matching Unlike other deductive database implementations such as LDL [8] CORAL [22], Aditi [26] Glue Nail [9] and LogicBase [12] the ROL system keeps intensional information derived with rules in the ROL space if the memory space permits. However, intensional information is not made persistent as extensional information in ROL. Therefore, the result to the user query may ....

.... Z] X : employee[salary Z] X : student[age Z] X : student[address Z] 6 Comparison with Related Systems In this section, we discuss major differences and similarities between the ROL system and other related deductive object base systems in terms of query processing. As LDL[8] and CORAL[22], ROL is a memory based single user system but supports persistent schema, facts and rules directly. ROL has a memory manager which manages the memory space. If the memory space is used up, the memory manager will remove the least recently used ones. Neither LDL nor CORAL support this. Besides, ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The coral deductive system. VLDB Journal, 3(2):161--210, January 1994.

....to or perform relatively poorly on others. ROL automatically decides which mechanism to use without user s intervention, based on the nature of the query and its knowledge about the data in the object database. 5.1. Matching Unlike other deductive database implementations such as LDL [10] CORAL [24], Aditi [30] Glue Nail [11] and LogicBase [14] ROL keeps intensional information derived with rules in the main memory if the memory space permits. However, intensional information is not made persistent as extensional information in ROL. Therefore, the result to the user query may already be ....

....[ sno Z] DESIGN AND IMPLEMENTATION OF THE ROL SYSTEM 19 O : course [ name Z] 6. Comparison with Related Systems In this section, we discuss major di erences and similarities between ROL and other related deductive object database systems in terms of query processing. As LDL[10] and CORAL[24], ROL is a memory based single user system but supports persistent schema, facts and rules directly. ROL has a memory manager which manages the memory space. If the memory space is used up, the memory manager will remove the least recently used ones. Neither LDL nor CORAL support this. Besides, ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161-210, 1994.

....There are several open issues which we still need to address. We intend to investigate how to incorporate update constructs into the language to make it a complete database language. We are also investigating how to efficiently implement the language based on the techniques used in LDL [12] CORAL [23] and Atlas [27] Acknowledgments The author wish to thank the anonymous referees for their detailed comments and suggestions which help to improve the quality and accuracy of the paper. This work was supported by the Natural Sciences and Engineering Research Council of Canada. ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The coral deductive system. VLDB Journal, 3(2):161--210, January 1994.

.... Results To confirm the theoretical work different implementations of typical algorithms are compared: ffl an F logic program with computed methods running on an XSB prototype using a straightforward (naive) implementation of the presented evaluation, ffl a Datalog program running on Coral [13] using two stategies: the system generated magic set optimization and a top down technique called Pipelining, ffl an F logic program without computed methods on the system Florid [7] working seminaive bottom up. To enable a comparison the examples do not rely on the object oriented constructs ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL deductive system. The VLDB Journal, 3(2):161--210, Apr. 1994.

.... To test the performance of computed methods the length of a chain in a binary relation of links (abstract variant of the introductory example) was implemented in three different versions: as a computed method on a prototypical Prolog interpreter for F logic, as a Datalog version running on Coral [5] using Coral s standard magic set technique, and as F logic program running on Florid [1] under seminaive bottom up evaluation. For different numbers of link facts and various lengths of the chain the following times in Cpu seconds were consumed: facts length computed methods Coral: magic set ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL deductive system. The VLDB Journal, 3(2):161--210, Apr. 1994.

....builds on this multi paradigm foundation, integrating programming features from imperative, logic, and constraint paradigms. Basically, Melody programs can be thought of as having an imperative skeleton, with logic elements fleshing out the structure (this is similar in a number of ways to CORAL [27], though CORAL does not unify C and logic at the model level) The imperative skeleton acts by evaluating constraints and triggers and executing transition actions. Using OSM equivalent statements as explained above, a transition action can be thought of as a sequence of statements, with ....

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri, "The CORAL Deductive System," VLDB Journal, vol. 3, no. 2, pp. 161-210, Boxwood Press, Pacific Grove, California, 1994.

....to support unknown values that are common in database applications. Indeed, it is not clear how to perform deduction when there are unknown values in a deductive 29 database. We are currently investigating how to efficiently implement the language based on the techniques used in LDL [12] CORAL [24] and Atlas [27] The author wishes to thank the anonymous referees for their detailed comments and suggestions which have significantly improved the quality and accuracy of the paper. This work was supported by the Natural Sciences and Engineering Research Council of Canada. ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The CORAL deductive system. VLDB Journal, 3(2):161--210, 1994.

....specific for deductive databases have been the main focus of extensive research [4, 5, 7, 8, 23, 25, 42, 49, 57, 55] and a number of deductive database systems or prototypes based on Datalog have been developed and reported. These include Nail [41] LOLA [15] Glue Nail [14] XSB [50] CORAL [46], Aditi [58] LogicBase [22] Declare SDS [26] etc. See [45] for a survey of these deductive database systems. However, deductive databases based on Datalog only provide inexpressive flat structures and cannot directly support complex values such as nested tuples and sets common in novel database ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161--210, 1994.

....systems such as Nail [16] LOLA [6] Glue Nail [5] XSB [20] Aditi [21] LogicBase [7] Declare SDS [8] etc. only support flat relations which are found inappropriate for advanced applications. A few deductive database systems that support data with complex structures such as LDL [3] and CORAL [18] are only implemented as memory based systems and do not support direct access of data with complex structures [12] For these reasons, we have developed a novel deductive database system called Relationlog at the University of Regina, Canada. Unlike existing deductive database systems, the ....

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The CORAL Deductive System. VLDB Journal, 3(2):161--210, 1994.

.... The field of deductive databases is mainly the story of query evaluation algorithms for recursive rules using database techniques [1] Among these, the many variants of the magic set technique were the most successful and are now a standard component of any deductive database system (e.g. [7, 9]) An important reason for the success of the magic set technique was the claim that it is as least as efficient as the top down evaluation known from logic programming. An often cited formalization of this has been proven by Ullman in a paper called Bottom Up beats Top Down for Datalog ....

....symbols are excluded) ffl The rules contain no negation, i.e. they are of the form A B 1 ; Bm , where A and the B i are atoms. ffl Every variable of the head literal A must appear also in a body literal B i (allowedness range restriction) Of course, modern systems such as CORAL [7] allow more general Prolog rules, but the magic set technique was originally developed for the above class of programs, and we wish to avoid unnecessary complications here and concentrate on the efficiency problem. For simplicity, we assume that the given query Q is a single literal. This is no ....

[Article contains additional citation context not shown here]

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL deductive system. The VLDB Journal, 3:161--210, 1994.

No context found.

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. In The VLDB Journal, 3:161-- 210, 1994.

No context found.

R. Ramakrishnan, D. Srivastava, S. Sudarshan, and P. Seshadri. The CORAL Deductive System. In The VLDB Journal, 3:161-- 210, 1994.

No context found.

Raghu Ramakrishnan, Divesh Srivastava, S. Sudarshan, and Praveen Seshadri. The CORAL Deductive System. VLDB Journal: Very Large Data Bases, 3(2):161 -- 210, 1994.

No context found.

R. Ramakrishnan, D. Srivastava, S. Sudarshan, P. Seshadri. The CORAL deductive system. VLDB Journal 3,2 (1994) 161-210.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC