H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the IEEE ICDE Conf., pages 336--344, April 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....Of these, the nested loop join applies no reduction criterion. Table 1 gives an overview of research work categorized by the above criteria. Leung and Muntz [LM92] describe a partition based algorithm in a mulJoin Reduction criterion technique TT VT VT EA Sort Merge [LM93] SS93] this work [GS91] GS91] GS91] partition based [SE96] SE96] SSJ94] LM92] Nested Loop Incremental this work Table 1: Previous Work on Temporal Joins tiprocessor environment. Soo et al. SSJ94] introduce a partition based algorithm supporting valid time. Son and Elmasri [SE96] present partition based ....

....the nested loop join applies no reduction criterion. Table 1 gives an overview of research work categorized by the above criteria. Leung and Muntz [LM92] describe a partition based algorithm in a mulJoin Reduction criterion technique TT VT VT EA Sort Merge [LM93] SS93] this work [GS91] GS91] GS91] partition based [SE96] SE96] SSJ94] LM92] Nested Loop Incremental this work Table 1: Previous Work on Temporal Joins tiprocessor environment. Soo et al. SSJ94] introduce a partition based algorithm supporting valid time. Son and Elmasri [SE96] present partition based temporal ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the IEEE Conference on Data Engineering, pages 336--344, Los Alamitos, CA, USA, April 1991.

....temporal indices. This work was partially supported by NSF (IIS 9907477, EIA9983445) and the Department of Defense. 1 Introduction A temporal record has a key, some attributes and a time interval during which the record is valid. Temporal join predicates may involve the key and or time spaces [9]. Examples include the T Join (join two records if their intervals intersect) the E Join (join two records in their keys are equal) and the TE Join (keys are equal and intervals intersect) Due to large volume of temporal data, a user may be interested in a portion rather than the whole ....

....E Join and TE Join (in short, GT , GE and GTE Joins) Note that the plain T Join, E Join and TE Join are then special cases when the query range r is the whole key space and the query interval i is the whole time space. Temporal join research has focused on non indexed algorithms for plain joins [9, 15, 18, 23, 19, 20, 26] and typically involves a sequential scan of both relations. This is prohibitive for general temporal joins. Instead, an indexing scheme is beneficial since it can quickly direct the join towards the objects of interest. With an index present in each joined relation, a straightforward approach is ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev, "Query Processing Algorithms for Temporal Intersection Joins", Proc. of ICDE, 1991.

....defined this set of temporal joins, we show how all previously defined operators are related to this taxonomy of temporal joins. The previous operators considered include Cartesian product, JOIN, EQUIJOIN, NATURAL JOIN, and TIME JOIN [CC87, CC93] TE JOIN, TE OUTERJOIN, and EVENT JOIN [SG89, GS91, Seg93, SE96] and those based on Allen s [All83] interval relations (cf. LM90, LM92b, NA93] We show that many of these operators incorporate less restrictive predicates or use specialized attribute semantics, and thus are variants of one of the taxonomic joins. 2.1 Temporal Join Definitions ....

....at each point in time. When operating on interval stamped data, this semantics corresponds to an intersection: the result will be valid during those times when contributing tuples from both input relations are valid. The temporal Cartesian product was first defined by Segev and Gunadhi [SG89, GS91] This operator was termed the Time join and the abbreviation T join was used. Clifford and Croker [CC93] defined a Cartesian product operator that is a combination of the temporal Cartesian product and the temporal outerjoin, to be defined shortly. 2.3 Theta Join Like the conventional ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the IEEE Conference on Data Engineering, Kobe, Japan, 1991.

....comes from the field of temporal databases. Salzberg and Tsotras provide an excellent overview of temporal indexing methods developed in the community over the past decade [27] Query optimization and cost models for temporal databases have been studied in early 90s by Gunadhi and Segev [13, 14], and their study was later continued by Soo, Snodgrass, Jensen and Slivinskas [33, 30] Their research was primarily concentrated on designing efficient algorithms for relational operations on temporal databases which minimized I O costs. These results, however, are not directly applicable to the ....

H. Gunadhi, A. Segev (1991) Query Processing Algorithms for Temporal Intersection Joins, in Proc. ICDE'91, pp. 336-344.

....Several temporal join algorithms have been presented in the literature. Table 1 classifies temporal joins by the join algorithm class and additional information about distribution or organisation of the data that is used by the algorithm during the join process. Temporal joins are categorised in [GS91]. Different join algorithms, taking physical organisation and indexing of the data into account, are presented and their cost analysed. A sort merge based algorithm for time oriented data is presented in [PJ99] An incremental approach is used to minimize the cost for updating temporal joins in ....

....space required to reduce disk I O. A partition based temporal join is presented in [SSJ94] Sampling is used to approximate the distribution of 1 Join Additional Information Techniques Used Algorithm Sorting Sampling Index Histogram Append only Parallel Class database processing (sort) merge [GS91] [GS91] PJ99] SS99] SS99] nested loop [RF93] partition based [SSJ94] SE96] this paper [LM92] Table 1: Classification of previous work on temporal joins the data and to compute the partitioning. The Time Index is used in [SE96] for a temporal join. The Time Index stores temporal entities ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proc. 7th ICDE, pages 336--344, April 1991.

....a time interval representing the record s time validity. Various temporal join predicates have been proposed ( OS95] Examples include the T Join (join two records if their intervals intersect) and the more general TE Join (join two records if their keys are equal and their intervals intersect) [GS91]) Here we examine a generalization of the TE Join (the GTEJ join) which also specifies that joined records must have their keys in range r while their intervals should intersect interval i. Such joins allow the user to concentrate on interesting rectangles in the key time space. As an example, ....

....one maintaining the graduate students at UC Riverside and the other the IBM summer interns. A GTEJ query is: find the students at UC Riverside with last names starting with B that during 1995 2000 were also working at IBM . Previous temporal join research has focused on non indexed joins ([GS91, LM93, RF93, SSJ94, RS96, SE96, Zur97]) The common approach is to scan both relations and select the records that need to be joined. Even though a plain scan of both relations will take advantage of sequential I O, it becomes prohibitive when the temporal relation sizes are large (as in a data warehouse) and the join selectivity is ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev, "Query Processing Algorithms for Temporal Intersection Joins", Proc. of ICDE, pp. 336-344, 1991.

....Universitat Marburg, Germany. seeger Mathematik.Uni Marburg.de 1 join predicates have been proposed ( OS95] Examples include the T Join (join two records if their intervals intersect) and the more general TE Join (join two records if their keys are equal and their intervals intersect) [GS91]) Here we examine a generalization of the TE Join (the GTE Join) which also specifies that joined records must have their keys in range r while their intervals should intersect interval i. Such joins allow the user to concentrate on interesting rectangles in the key time space. As an example, ....

....one maintaining the graduate students at UC Riverside and the other the IBM summer interns. A GTE Join query is: find the students at UC Riverside with last names starting with B that during 1995 2000 were also working at IBM . Previous temporal join research has focused on non indexed joins ([GS91, LM93, RF93, SSJ94, RS96, SE96, Zur97]) The common approach is to scan both relations and select the records that need to be joined. Even though a plain scan of both relations will take advantage of sequential I O, it becomes prohibitive when the temporal relations are large (as in a data warehouse) and the join selectivity is high ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev, "Query Processing Algorithms for Temporal Intersection Joins", Proc. of ICDE, pp. 336-344, 1991.

....deal with di erent operators and di erent properties of temporal data. As in relational databases, ecient evaluation of the join operator is crucial for the overall performance of the system. Several temporal join algorithms have been presented in the literature. Temporal joins are categorised in [GS91] taking the physical organisation and indexing of the data into account. A sort merge based algorithm for time oriented data is presented in [PJ99] An incremental approach is used to minimize the cost for updating temporal joins in append only databases. In [RF93] several re nements of the ....

....where each tuple t is stamped with a time interval [t:t s , t:t e ] indicating the range of time where the tuple is valid. The time attributes associated with temporal relations can refer to either the valid time or the transaction time. A classi cation of temporal joins is given in [GS91]. In this paper, we focus on evaluation of the valid time time join. The time join nds tuples with intersecting time intervals without specifying a join predicate on non time attributes. The time equi join also requires equality on non time attributes. Our approach can also be used for time equi ....

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proc. 7th ICDE, pages 336-344, April 1991.

.... been defined, including the time join, event join, TEouterjoin [SG89] contain join, contain semijoin, intersect join, overlap join [LM91a] and containsemijoin [LM92] Refinements to the nested loops algorithm were proposed by Gunadhi and Segev to evaluate several temporal join variants [SG89, GS91] This work assumed that temporal data was append only, i.e. tuples are inserted in timestamp order into a relation, and once inserted into a relation are never deleted. With the append only assumption, a new access path, the append only 12 ....

....p ( samples r ) return cachePages Figure 9: Tuple Cache Size Estimation tree, was developed that provides a temporal index on the relation. Simple extensions to sort merge were also considered where again tuples were assumed to be inserted into a relation in timestamp order [SG89, GS91] Leung and Muntz extended this work to accommodate additional temporal join predicates, mainly those defined by Allen [All83] and to incorporate various ascending and descending sort orders on either valid start or valid end time [LM90] Simply stated, our work differs from most previous work ....

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the 7th International Conference on Data Engineering, Kobe, Japan, 1991.

.... joins have been defined, including the time join, eventjoin, TE outerjoin [21] contain join, contain semijoin, intersect join, overlap join [17] and contain semijoin [16] Refinements to the nested loops algorithm were proposed by Gunadhi and Segev to evaluate several temporal join variants [21, 7]. This work assumed that temporal data was append only, i.e. tuples are inserted in timestamp order into a relation, and once inserted into a relation are never deleted. With the append only assumption, a new access path, the append only tree,was developed that provides a temporal index on the ....

....are never deleted. With the append only assumption, a new access path, the append only tree,was developed that provides a temporal index on the relation. Simple extensions to sort merge were also considered where, again, tuples were assumed to be inserted into a relation in timestamp order [21, 7]. Leung and Muntz extended this work to accommodate additional temporal join predicates, mainly those defined by Allen [1] and to incorporate various ascending and descending sort orders on either validstart or valid end time [15] Simply stated, our work differs from most previous work in that ....

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the 7th International Conference on Data Engineering, Kobe, Japan, 1991.

....by the possibility. Since the uncertainty is not explicitly represented, and the possibility is the upper bound, the satisfaction degree obtained using this method may, at times, be counter intuitive. Several interval based join algorithms were proposed in the context of temporal databases [5] or ordinary relational databases [4] The temporal join algorithms join tuples that have overlapping time periods. The band join algorithm in [4] joins a tuple r with a tuple s if the join attribute value of s is within a pre specified neighborhood of that of r. These algorithms are not ....

H.Gunadhi and A. Segev. Query processing algorithms for temporal intersection joins. In IEEE Int'l Conf. on Data Engineering, Kobe, Japan, 1991.

....two dimensional range searching presented in [16] and the algorithm for batch two dimensional range queries presented in [10] one can obtain an asymptotically optimal algorithm for the interval join problem. On the other hand, the approaches in the temporal database literature for interval join [12,24] are based on heuristics and do not offer good asymptotic worst case bounds. In this paper, we present an extremely simple and elegant optimal solution for this problem. After an initial sort of the two input relations, the algorithm uses a single scan of the sorted relations, in which it ....

H. Gunadhi & A. Segev, "Query Processing Algorithms for Temporal Intersection Joins," Proc. IEEE Int'l Conf. on Data Engg. (1991).

....partition based, nested loop, and incremental techniques [ME92] Of these, the nested loop join applies no reduction criterion. Table 1 gives an overview of research work categorized by the above criteria. Join Reduction criterion technique TT VT VT EA Sort Merge [LM93] SS93] this work [GS91] GS91] GS91] partition based [LM92] SSJ94] Nested Loop Incremental this work Table 1: Previous Work on Temporal Joins Leung and Muntz [LM92] describe a partition based algorithm in a multiprocessor environment. Soo et al. SSJ94] introduce a partitioning based algorithm supporting valid time. ....

....partition based, nested loop, and incremental techniques [ME92] Of these, the nested loop join applies no reduction criterion. Table 1 gives an overview of research work categorized by the above criteria. Join Reduction criterion technique TT VT VT EA Sort Merge [LM93] SS93] this work [GS91] GS91] GS91] partition based [LM92] SSJ94] Nested Loop Incremental this work Table 1: Previous Work on Temporal Joins Leung and Muntz [LM92] describe a partition based algorithm in a multiprocessor environment. Soo et al. SSJ94] introduce a partitioning based algorithm supporting valid time. Segev ....

[Article contains additional citation context not shown here]

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the IEEE Conference on Data Engineering, pages 336--344, Los Alamitos, CA, USA, April 1991.

....of monotonically increasing size. For an algebra to utilize this approach, incremental forms of the operators are required (c.f. 92] 6.1. 2 Temporal Joins A wide variety of binary joins have been considered, including time join and time equijoin (TEjoin) 42] event join and TE outerjoin [71], contain join, contain semijoin and intersect join [129] and temporal natural join [199] The various algorithms proposed for these joins have generally been extensions to nested loop or merge joins that exploit sort orders or local workspace, as well as hash joins. More work is necessary to ....

Gunadhi, H., and Segev, A., "Query processing algorithms for temporal intersection joins," in Proceedings of the International Conference on Data Engineering, Kobe, Japan, 1991.

....of reference in sequel, we use the term join to refer exclusively to the time join, and the term relation to mean temporal relation, unless otherwise stated. 2.2. Partitioned based Temporal Joins Initial work on temporal joins focused on refinements of the conventional nested loops algorithm [2, 3, 9]. These algorithms exploit the sort order of the relations to avoid full scan of the inner and or outer relations. Partition based algorithms for temporal join proposed in the literature can be classified into the following three types: Static Partitioning [4] In this method, R and S are ....

H. Gunadhi and A. Segev. Query processing algorithms for temporal intersection joins. In Proceedings of the Seventh International Conference on Data Engineering, pages 336--344, Kobe, Japan, April 1991.

....warehouse has a very strict sense of what time period has been captured, and the user is only permitted to access this data in a relatively small number of discrete snapshots. As such, access structures are not necessary. Query optimization in temporal databases is another area of concern [26] [27], 28] 34] 40] We will consider this issue somewhat on presentation of our method. Other work has looked at how to impose constraints on a temporal database, e.g. salaries must never decrease. 39] We will not consider these ideas here. One way of interpreting a snapshot is as a ....

H. Gunadhi and A. Segev, "Query Processing Algorithms for Temporal Intersection Joins", Proceedings of the International Conference on Data Engineerin, 1991.

.... timestamp represents the positions in time at which the tuple(object) was true . The associated query language is extended with predicates that access the timestamp of the tuples (objects) There has been some work on query optimization based on such models[GS89a, LM93, NG93] For instance [GS91, LM93] propose efficient stream access techniques of processing various types of temporal joins , and [GS89b] proposes an optimization framework for temporal data based on such techniques. Our approach to sequences presented in this paper takes a strongly positional view of sequences, as ....

Himawan Gunadhi and Arie Segev. Query processing algorithms for temporal intersection joins. In Proceedings of the International Conference on Data Engineering, 1991.

.... the expressive power of a database while query optimization is a critical issue in practicality of a temporal database management system(TDBMS) So far a number of strategies to process temporal operators for temporal join, which is one of the most expensive operations in TDBMS, were proposed[GS91, SSJ94, LOT94] In general, join evaluation algorithms fall into three basic categories, nested loops, sort merge, and partition based. Among them a partition based approach, which has a great potential for effective temporal join handling, is our main concern. However, without some replications, ....

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Join. In Proceeding of the 7th ICDE, 1991.

No context found.

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the IEEE ICDE Conf., pages 336--344, April 1991.

No context found.

H. Gunadhi and A. Segev. Query Processing Algorithms for Temporal Intersection Joins. In Proceedings of the IEEE Conference on Data Engineering, pages 336--344, Los Alamitos, CA, USA, April 1991.

No context found.

Himawan Gunadhi and Arie Segev. Query processing algorithms for temporal intersection joins. In ICDE 1991.

No context found.

H. Gunadhi and A. Segev, \Query Processing Algorithms for Temporal Intersection Joins", Proc. of ICDE, 1991.

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