Google Inc.

Demandtec, Inc.

Pivotal Software, Inc.

M-factor, Inc.

Chad Whipkey

Phillip Dennis Delurgio

Michael Alexander Shwe

Steven John Peter Hillion

Dirk Manfred Beyer

James Drew

Bart Samwel

Nathaniel Keith Folkert

Florian Michael Waas

Sivaramakrishnan Narayanan

Brian C Hagenbuch

Brian Frederick Babcock

An apparatus for conducting a hierarchical join on a hierarchically clustered data stream is disclosed. The apparatus may include a memory storing query conditions that each determine whether a Structured Query Language query is suitable for performing a non-buffering read operation on a hierarchical data storage. A processor in communication with the memory may be configured to receive a Structured Query Language ('SQL') query from a client device and evaluate the SQL query for one or more of the query conditions stored in the memory. When the processor determines that the SQL query satisfies one or more of the query conditions, the processor may then perform a non-buffering read operation on the hierarchical data storage to obtain a hierarchically clustered data stream. The hierarchically clustered data stream may be evaluated to obtain a plurality of record results, which may be transmitted to the client device.

Efficient top-down hierarchical join on a hierarchically clustered data stream

A method of analyzing the performance of a query optimizer includes identifying an event trigger. A reproduction object characterizing operational parameters of the customer computer at the time of the event trigger is populated. The reproduction object is transported from the customer computer to a test computer. The reproduction object is analyzed at the test computer to characterize the performance of a query optimizer.

Apparatus and method for analyzing query optimizer performance

Methods and apparatuses for computing a variance between two business metrics is described. In one embodiment, the method computes a variance for each of a first set of activities based on the corresponding reference state of that activity, wherein the variance for an activity is the change in contribution for that activity between the first and second business metrics and with each of the first set of activities having a reference value. Furthermore, the method computes a variance for each of a second set of activities based on the corresponding start and end values of that activity with each of the second set of activities having a start and end value.

Method and apparatus for creating compound due-to reports

Methods and apparatuses for computing a variance for the difference between two business metrics. In one embodiment, the method accesses a response model and a plurality of activities with start and end values for each of the plurality of activities. Furthermore, the method computes a variance for the difference between the first and second business metrics for each of the plurality of activities using the response model by setting that activity to one of the corresponding starting and ending values and setting others of the plurality of activities to the value state opposite of that activity, wherein the variance for an activity is the change in contribution for that activity between the start and end sales volumes.

Method and apparatus for creating due-to reports for activities that may not have reference value

Methods and apparatuses for decomposing a business metric based on a plurality of activities and a response model are described. In one embodiment, the method accesses the response model and the plurality of activities, the plurality of activities each having a reference and executed value. The method computes a contribution to the business metric based on setting one of the plurality of activities to one of the corresponding reference and executed value and setting the other activities to the value state opposite of that activity. Furthermore, the method computes each of the contributions independent of the response model type.

Method and apparatus for configurable model-independent decomposition of a business metric

Methods and apparatuses for predicting set of multi-dimensional dependent data and non-measurable data from a set of multi-dimensional historical dependent and causal data are described. In one embodiment, the method comprises receiving input data that comprises multi-dimensional historical dependent data and causal data and anticipated activity data, determining a set of multi-dimensional predicted dependent data using a predictive model and the input data, creating non-measurable data based on the set of multi-dimensional predicted dependent data and the input data.

Growing community of inventors

Foster City, CA, United States of America

Chad Whipkey