US 2026/0310631 A1Published application

Horizontally Scalable Multi-Dimensional Analysis Architecture

AssigneeBlue Yonder Group, Inc.

InventorsSteven R. Weiss, Steve Pawlish

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system and method are disclosed including a distributed computer network having one or more application servers and one or more data nodes. The one or more data nodes each include a shard of data representing a measure. The one or more application servers receive a measure model and an edit to the measure. The one or more application servers append the edit to a register and generate a calculation sequence. The one or more application servers further disaggregate one or more values to a shard of data on the one or more data nodes based, at least in part, on the edited measure and calculate one or more unedited measures based, at least in part, on the disaggregation.

Claims20

Dependent claims

A system comprising:

a distributed computer network comprising one or more application servers in communication with one or more data nodes, the one or more data nodes storing multi-dimensional data as shards, the one or more application servers comprising a distributor, a calculation engine, an aggregator, a disaggregator, a queue, and memory;
wherein the calculation engine is configured to:
receive a measure model comprising measures, rule sets, and flexibility parameters;
receive an edit to a first measure of the measures;
append the edit to the queue;
generate a calculation sequence from the rule sets based on pending edits in the queue;
determine a disaggregation type associated with the first measure;
cause the disaggregator, through the distributor, to disaggregate one or more values to one or more of the shards on the one or more data nodes based on the edit and the disaggregation type;
and
calculate one or more unedited measures of the measures based on the disaggregation and the calculation sequence.

7 more elements

A method comprising:

receiving, by one or more application servers in communication with one or more data nodes storing multi-dimensional data as shards, a measure model comprising measures, rule sets, and flexibility parameters, the one or more application servers comprising a distributor, a calculation engine, an aggregator, a disaggregator, a queue, and memory;
receiving, by the calculation engine, an edit to a first measure of the measures;
appending, by the calculation engine, the edit to the queue;
generating, by the calculation engine, a calculation sequence from the rule sets based on pending edits in the queue;
determining, by the calculation engine, a disaggregation type associated with the first measure;
causing, by the calculation engine, the disaggregator through the distributor to disaggregate one or more values to one or more of the shards on the one or more data nodes based on the edit and the disaggregation type;
and
calculating, by the calculation engine, one or more unedited measures of the measures based on the disaggregation and the calculation sequence.

5 more elements

A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of one or more application servers in communication with one or more data nodes storing multi-dimensional data as shards, cause the one or more application servers to perform operations, the one or more application servers comprising a distributor, a calculation engine, an aggregator, a disaggregator, a queue, and memory, the operations comprising:

receiving a measure model comprising measures, rule sets, and flexibility parameters;
receiving an edit to a first measure of the measures;
appending the edit to the queue;
generating a calculation sequence from the rule sets based on pending edits in the queue;
determining a disaggregation type associated with the first measure;
causing the disaggregator, through the distributor, to disaggregate one or more values to one or more of the shards on the one or more data nodes based on the edit and the disaggregation type;
and
calculating one or more unedited measures of the measures based on the disaggregation and the calculation sequence.

5 more elements

Dependencies are inferred from the claim text.