US 11,769,065 B2Granted

Distributed system generating rule compiler engine by determining a best matching rule based on concrete parameterization with declarative rules syntax

AssigneeJulius Technologies LLC

InventorYadong Li

Life of this patent

Estimating the term…

Loading patent…

Abstract

An output rule specified via a distributed system execution request data structure for a requested calculation is determined, and a current rule is initialized to the output rule. A rule lookup table data structure is queried to determine a set of matching rules, corresponding to the current rule. The best matching rule is selected. A logical dependency graph (LDG) data structure is generated by adding LDG nodes and LDG edges corresponding to the best matching rule, precedent rules of the best matching rule, and precedent rules of each precedent rule. An execution complexity gauge value and a set of distributed worker processes are determined. The LDG data structure is divided into a set of subgraphs. Each worker process is initialized with the subgraph assigned to it. Execution of the requested calculation is coordinated and a computation result of the LDG node corresponding to the output rule is obtained.

Claims41

Dependent claims

A rule compiler apparatus for generating a distributed system, comprising:

a memory;
a processor in communication with the memory, and configured to issue a plurality of processing instructions stored in the memory, wherein the processor issues instructions to:
(a) obtain, via the processor, a distributed system execution request data structure;
(b) determine, via the processor, an output rule structured to comprise a declaration implementing polymorphic parameterization via a declarative rules syntax and a set of rule parameters specifying concrete parameterization for the declaration in accordance with the declarative rules syntax associated with the output rule specified via the distributed system execution request data structure for a requested calculation, wherein a current rule is initialized to the output rule;
(c) query, via the processor, a rule lookup table data structure to determine a set of matching rules, corresponding to the current rule, based on an identification tuple associated with the current rule;
(d) select, via the processor, a best matching rule from the set of matching rules, wherein a set of rule parameters associated with the best matching rule best corresponds to a set of rule parameters associated with the current rule;
(e) generate, via the processor, a logical dependency graph data structure by adding logical dependency graph nodes and logical dependency graph edges corresponding to the best matching rule, precedent rules of the best matching rule, and precedent rules of each of the precedent rules of the best matching rule;
(f) determine, via the processor, an execution complexity gauge value associated with the generated logical dependency graph data structure;
(g) determine, via the processor, a set of distributed worker processes to be utilized to execute the requested calculation;
(h) divide, via the processor, the generated logical dependency graph data structure into a set of subgraphs based on the execution complexity gauge value, wherein each worker process in the set of distributed worker processes is assigned a subgraph from the set of subgraphs;
(i) initialize, via the processor, each respective worker process in the set of distributed worker processes with the subgraph assigned to the respective worker process;
(j) coordinate, via the processor, execution of the requested calculation by worker processes in the set of distributed worker processes, wherein each respective worker process calculates results for logical dependency graph nodes in the subgraph assigned to the respective worker process;
and
(k) obtain, via the processor, a computation result of a logical dependency graph node from the logical dependency graph nodes corresponding to the output rule, from the worker process in the set of distributed worker processes that is assigned the subgraph that contains the logical dependency graph node corresponding to the output rule.

11 more elements

A rule compiler processor-readable non-transitory physical medium for generating a distributed system storing processor-issuable instructions to:

(a) obtain, via the processor, a distributed system execution request data structure;
(b) determine, via the processor, an output rule structured to comprise a declaration implementing polymorphic parameterization via a declarative rules syntax and a set of rule parameters specifying concrete parameterization for the declaration in accordance with the declarative rules syntax associated with the output rule specified via the distributed system execution request data structure for a requested calculation, wherein a current rule is initialized to the output rule;
(c) query, via the processor, a rule lookup table data structure to determine a set of matching rules, corresponding to the current rule, based on an identification tuple associated with the current rule;
(d) select, via the processor, a best matching rule from the set of matching rules, wherein a set of rule parameters associated with the best matching rule best corresponds to a set of rule parameters associated with the current rule;
(e) generate, via the processor, a logical dependency graph data structure by adding logical dependency graph nodes and logical dependency graph edges corresponding to the best matching rule, precedent rules of the best matching rule, and precedent rules of each of the precedent rules of the best matching rule;
(f) determine, via the processor, an execution complexity gauge value associated with the generated logical dependency graph data structure;
(g) determine, via the processor, a set of distributed worker processes to be utilized to execute the requested calculation;
(h) divide, via the processor, the generated logical dependency graph data structure into a set of subgraphs based on the execution complexity gauge value, wherein each worker process in the set of distributed worker processes is assigned a subgraph from the set of subgraphs;
(i) initialize, via the processor, each respective worker process in the set of distributed worker processes with the subgraph assigned to the respective worker process;
(j) coordinate, via the processor, execution of the requested calculation by worker processes in the set of distributed worker processes, wherein each respective worker process calculates results for logical dependency graph nodes in the subgraph assigned to the respective worker process;
and
(k) obtain, via the processor, a computation result of a logical dependency graph node from the logical dependency graph nodes corresponding to the output rule, from the worker process in the set of distributed worker processes that is assigned the subgraph that contains the logical dependency graph node corresponding to the output rule.

9 more elements

A processor-implemented rule compiler method for generating a distributed system, comprising executing processor-implemented instructions to:

(a) obtain, via the processor, a distributed system execution request data structure;
(b) determine, via the processor, an output rule structured to comprise a declaration implementing polymorphic parameterization via a declarative rules syntax and a set of rule parameters specifying concrete parameterization for the declaration in accordance with the declarative rules syntax associated with the output rule specified via the distributed system execution request data structure for a requested calculation, wherein a current rule is initialized to the output rule;
(c) query, via the processor, a rule lookup table data structure to determine a set of matching rules, corresponding to the current rule, based on an identification tuple associated with the current rule;
(d) select, via the processor, a best matching rule from the set of matching rules, wherein a set of rule parameters associated with the best matching rule best corresponds to a set of rule parameters associated with the current rule;
(e) generate, via the processor, a logical dependency graph data structure by adding logical dependency graph nodes and logical dependency graph edges corresponding to the best matching rule, precedent rules of the best matching rule, and precedent rules of each of the precedent rules of the best matching rule;
(f) determine, via the processor, an execution complexity gauge value associated with the generated logical dependency graph data structure;
(g) determine, via the processor, a set of distributed worker processes to be utilized to execute the requested calculation;
(h) divide, via the processor, the generated logical dependency graph data structure into a set of subgraphs based on the execution complexity gauge value, wherein each worker process in the set of distributed worker processes is assigned a subgraph from the set of subgraphs;
(i) initialize, via the processor, each respective worker process in the set of distributed worker processes with the subgraph assigned to the respective worker process;
(j) coordinate, via the processor, execution of the requested calculation by worker processes in the set of distributed worker processes, wherein each respective worker process calculates results for logical dependency graph nodes in the subgraph assigned to the respective worker process;
and
(k) obtain, via the processor, a computation result of a logical dependency graph node from the logical dependency graph nodes corresponding to the output rule, from the worker process in the set of distributed worker processes that is assigned the subgraph that contains the logical dependency graph node corresponding to the output rule.

9 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/816,367Filed 12 Mar 2020 · us provisional application
Priority: 14 Mar 2019
Published: US 2020/0293916 A117 Sep 2020
Granted: 26 Sep 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification10 codes

G06N 5/025
Knowledge engineering; Knowledge acquisition: extracting rules from data
G06F 9/5066
Partitioning or combining of resources: algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs
G06F 9/5072
Partitioning or combining of resources: grid computing
H04L 67/10
Protocols: in which an application is distributed across nodes in the network

Family

Application62/818,318Us provisional application · filed 14 Mar 2019
Application62/892,085Us provisional application · filed 27 Aug 2019
Application62/894,001Us provisional application · filed 30 Aug 2019
US 2020/0293916 A1Pre-grant publication · published 17 Sep 2020
US 11,769,065 B2This patent

From the US filing history. Foreign family members aren’t listed.