With non-traditional application areas such as engineering design, image/voice data management, scientific/statistical applications, and artificial intelligence systems all clamoring for ways to store and efficiently process larger and larger volumes of data, it is clear that traditional database technology has been pushed to its limits. It also seems clear that no single database system will be capable of simultaneously meeting the functionality and performance requirements of such a diverse set of applications. In this paper we describe the initial design of EXODUS, an extensible database system that will facilitate the fast development of high-performance, applicationspecific database systems. EXODUS provides certain kernel facilities, including a versatile storage manager and a type manager. In addition, it provides an architectural framework for building application-specific database systems, tools to partially automate the generation of such systems, and libraries of software components (e.g., access methods) that are likely to be useful for many application domains.

No abstract

ChemInform is a weekly Abstracting Service, delivering concise information at a glance that was extracted from about 100 leading journals. To access a ChemInform Abstract of an article which was published elsewhere, please select a “Full Text” option. The original article is trackable via the “References” option.

No abstract

In this paper we &cuss the opnmtzatmn of multiple-relation multiple-disJunct queries in a relauonal database system Since optrmrzahon techniques for conjunctive (single disjunct) queries m relauonal databases are well known [Smtth75, Wong76, Selmger79, Yao79, Youssefi793, the natural way to evaluate a mulhple-drsJunct query was to execute each tiJunct mdependently [Bernstem 1, Kerschberg821 However, evaluahng each drsJunct Independently may be very mefficrent. In thus paper, we develop methods that merge two or more drsJuncts to form a term The advantage of mergmg drsJuncts to form terms hes m the fact that each term can be evaluated wnh a smgle scan of each relation that 1s present m the term. In add&on, the number of umes a Jam is performed will also be reduced when two or more drsJuncts are merged The critena for mergmg a set of CfisJtmcts wrll be presented PermIssion to copy wlthout fee all or part of this mate& IS granted provided that the copies are not made or dlstrtbuted for direct commerclal advantage, the ACM copyrlght notice and the title of the pubhcatlon and Its date appear, and notlce IS given that copymg IS by permIssIon of the Assoctatlon for Computmg Machmery To copy otherwlse, or to repubhsh, reqmres a fee and/or speck permlsslon 3 1988 ACM O-89791-263-2/88/0003/0263 $1 50 As we wdl see, the number of times each relatron m the query 1s scanned wrll be equal to the number of terms Thus, mmlmrzmg the number of terms wrll mmmize the number of scans for each relatron. We wtll formulate the problem of rmmmrzmg the number of scans as one of covermg a merge graph by a muumum number of complete merge graphs which are a resmcted class of carte&an product graphs. In general, the problem of mmunrzmg the number of scans 1s NP-complete. We present polynomral hme algonthms for spectal classes of merge graphs We also present a huensttc for general merge graphs. Throughout thrs paper, we wrll assume that no relahons have any mdrces on them and that we are only concerned wtth reducmg the number of scans for each relation present m the query What about relahons that have mdrces on them' It turns out that our performance memc of reducing the number of scans is beneficral even in the case that there are indrces. In lMuxaldarshna88] we demonstrate that when ophmtzmg smglerelahon mulnple-drsJunct queries, the cost (measured m terms of drsk accesses) may be reduced If all the IsJuncts are opumrzed together rather than mdrvrdually Thus, our algorithm for mrmmrzmg the number of terms 263

No abstract

No abstract

No abstract