On approximation-based query evaluation, expensive predicates and constraint objects.pdfVIP

下载本文档

0
0
约5.89万字
约 20页
2017-04-11 发布于江苏
举报
版权申诉

On approximation-based query evaluation, expensive predicates and constraint objects.pdf

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

On Approximation-based Query Evaluation, Expensive Predicates andConstraint ObjectsAlexander Brodskyyand X. Sean WangzDepartment of Information and Software Systems EngineeringGeorge Mason University, Fairfax, VA 22030fbrodsky,xywangg@isse.gmu.eduAbstractThe use of approximations of complex, possibly nested collection, objects can considerably im-prove the processing of queries with expensive predicates. The improvement is gained by (1) avoidingmuch of unnecessary evaluation of expensive predicates, and (2) enabling the use of a signicantlywider range of special-purpose algorithms, and indexing techniques working on approximations ofobjects. This paper develops a necessary query transformation infrastructure for such approximation-based query processing, in presence of extensible family of collection types, aggregation functions,indices and special purpose algorithms. The approximation-based ltering is formalized by the notionof a ltering graph, whose arcs describe, intuitively, under which conditions a ltering predicate canbe added before a predicate in a query. Next, two pre-processing mechanisms, the inverse groupingand approximation grouping are introduced to facilitate a direct retrieval of nested collection objectsby testing their internal components, their transformations or approximations. Query transformationrules corresponding the ltering graph and the groupings are presented using a variation of monoidcomprehension calculus. While the ltering graph and the groupings are powerful tools in optimiza-tion, the ltering graph must, in general, be provided the application programmer. However, thispaper introduces and studies a very expressive family of constraint objects, for which the lteringgraph and inverse and approximation groupings can be derived automatically. The paper also con-structs complex query transformation templates which are derived from the basic transformationrules for ltering graphs, and for the two pre-processing mechanisms.1 Intro