Visa is the payments technology that forms the backbone of the world's financial systems by handling more than 7 trillion dollars of payments annually and our data reflects how the world spends money. We will describe technical achievements we have made in the area of fraud and cover some open challenges in data science.

Insights based on detailed data on consumer behavior, product performance and marketplace behavior are driving innovation and competition in the internet space. We introduce Everest, a SQL-compliant data warehousing engine, based on a column architecture that we have built and deployed at Yahoo!. In contrast to commercially available engines, this massively parallel engine, based on commodity hardware, offers scale, flexibility, specialized analytic operations, and lower administrative & hardware costs. In this paper, we describe the business motivation and the software and deployment architecture of Everest. The engine is in production at Yahoo! since 2007 and currently manages over six petabytes of data.

The complexity of queries in relational DBMSs is increasing, particularly in the decision support area and interactive client sewer environments. This calls for a more powerful and flexible optimization of complex queries. H. Pirahesh et al. (1992) introduced query rewrite as a distinct query optimization phase mainly targeted to responding to this requirement. This approach has enabled us to extensively enrich the optimization rules in our system. Further, it has made it easier to incrementally enrich and adapt the system as need arises. Examples of such query optimizations are predicate pushdown, subquery and magic sets transformations, and decorrelating subquery. We describe the design and implementation of a rule engine for query rewrite optimization. Each transformation is implemented as a rule which consists of a pair of rule condition and action. Rules can be grouped into rule classes for higher efficiency, better understandability and more extensibility. The rule engine has a number of novelties in that it supports a full spectrum of control-from totally data driven to totally procedural. Furthermore, it incorporates a budget control scheme for controlling the resources taken for query optimization as well as guaranteeing the termination of rule execution. The rule engine and a suite of query rewrite rules have been implemented in Starburst relational DBMS prototype and a significant portion of this technology has been integrated into IBM DB2 Common Server relational DBMS.

We introduce a class of novel multiprocessor scheduling problems that arise in the optimization of SQL queries for

The goal of the Papyrus project [3] is to provide tools and services to enable the integration and parallelization of specialized data managers so that data-intensive applications can be constructed easily and efficiently. In our terminology, a data manager (DM) is a set of specialized methods that manage persistent data. A collection of functions defines the interface to a DM and provides the only means of accessing its persistent data.

The decreasing cost of computing makes it economically viable to reduce the response time of decision support queries by using parallel execution to exploit inexpen-sive resources. This goal poses the following query op-timization problem: Mzntmzze response ttme subject to constraints on throughput, which we motivate as the dual of the traditional DBMS problem, We address this novel problem in the context of Select-Project-Join queries by extending the execution space, cost model and search al-gorithm that are widely used in commercial DBItlSs. We incorporate the sources and deterrents of parallelism in the traditional execution space. We show that a cost model can predict response time while accounting for the new aspects due to parallelism, We observe that the response time optimization metric violates a fundamen-tal assumption in the dynamic programming algorithm that is the linchpin in the optimizers of most commer-cial DBMSS. We extend dynamic programming and show how optimization metrics which correctly predict response time may be designed. 1

This paper describes the Query Rewrite facility of the Starburst extensible database system, a novel phase of query optimization. We present a suite of rewrite rules used in Starburst to transform queries into equivalent queries for faster execution, and also describe the production rule engine which is used by Starburst to choose and execute these rules. Examples are provided demonstrating that these Query Rewrite transformations lead to query execution time improvements of orders of magnitude, suggesting that Query Rewrite in general—and these rewrite rules in particular—are an essential step in query optimization for modern database systems.

The goals of the Iris database management system are to enhance database programmer productivity and to provide generalized database support for the integration of future applications. Iris is based on an object and function model. Iris objects are typed but unlike other object systems, they contain no state. Attribute values, relationships and behavior of objects are modeled by functions. The Iris architecture efficiently supports the evaluation of functional expressions. The goal of the architecture is to provide a database system that is powerful enough to support the definition of functions and procedures that implement the semantics of the data model. An overview of the data model is provided, the architecture is described in detail, and implementation experience and usage of the system are discussed.>