Technologies for enabling rule engines to operate based on data chunking
This disclosure enables a cloud-based system that processes data requests in real-time using a serverless architecture. The system receives structured requests from a computing terminal, transforms those requests, and responsively runs statistical models to generate decision outputs. These outputs are stored as metadata for historical trend analysis and are also evaluated by a rule engine to produce a result. The system operates without pre-provisioned servers, using distributed compute nodes that scale automatically and complete processing within a time window. By combining parallel execution, dynamic configuration, and historical context, the system delivers fast, intelligent decisions with minimal infrastructure overhead.
1 . A system, comprising:
a serverless environment hosting a plurality of compute nodes distributively running a parsing logic, an execution logic, and a rule engine, wherein the rule engine accesses a plurality of rules;
a persistent storage containing a first serverless database, a second serverless database, and a third serverless database, wherein the first serverless database stores a plurality of executable files enabling a plurality of statistical models, wherein the second serverless database stores a plurality of configurations profiles, wherein the third serverless database stores a plurality of historical metadata enabling a plurality of historical trends; and
a computing terminal hosting an application program such that the application program generates a request in a first data format and sends the request in the first data format to the parsing logic, wherein the request has a plurality of attributes, wherein without real-time provisioning the plurality of compute nodes to handle the request during a time window:
a) the parsing logic receives the request in the first data format thereby initiating the time window, converts the request into a second data format, and sends the request in the second data format to the execution logic,
b) the execution logic receives the request in the second data format, queries the second serverless database to retrieve a subset of the plurality of configuration profiles based on selecting a subset of the plurality of attributes, queries the first serverless database to retrieve a subset of the plurality of executable files based on the subset of the plurality of configuration profiles, executes the subset of the plurality of executable files in parallel to respectively generate in parallel a plurality of outputs based on selecting a subset of the plurality of statistical models such that the plurality of outputs contains a plurality of indicators and a plurality of context identifiers identifying a subset of the plurality of historical trends associated with a plurality of requests submitted to the execution logic before the request, generate a first copy of the plurality of outputs in a metadata format, write the first copy in the metadata format into the third database data chunk-by-data chunk such that the first copy in the metadata format is stored in the plurality of historical metadata thereby updating the plurality of historical trends, generate a second copy of the plurality of outputs in the second data format, send the second copy in the second data format to the rule engine,
c) the rule engine receives the second copy in the second data format, applies at least one rule of the plurality of rules to the second copy such that a result is generated in the second data format and sent to the parsing logic, and
d) parsing logic receives the result in the second data format, converts the result from the second data format to the first data format and sends the result in the first data format to the application program thereby terminating the time window.
2 . The system of claim 1 , wherein the execution logic is logically interposed between the parsing logic and the rule engine.
3 . The system of claim 1 , wherein the execution logic is a software engine.
4 . The system of claim 1 , wherein the first serverless database is a NoSQL database.
5 . The system of claim 1 , wherein the second serverless database is a NoSQL database.
6 . The system of claim 1 , wherein the third serverless database is a NoSQL database.
7 . The system of claim 1 , wherein at least two of the first serverless database, the second serverless database, and the third serverless database are a NoSQL database.
8 . The system of claim 1 , wherein the plurality of executable files is a plurality of script files.
9 . The system of claim 1 , wherein the plurality of configurations profiles is stored in a structured data format or a delimited data format.
10 . The system of claim 1 , wherein the plurality of historical metadata is stored in a structured data format or a delimited data format.
11 . The system of claim 1 , wherein the plurality of historical trends is for about seven years or less.
12 . The system of claim 1 , wherein the time window is about five seconds or less.
13 . The system of claim 1 , wherein the plurality of outputs is output or stored in a structured data format or a delimited data format.
14 . The system of claim 1 , wherein the plurality of outputs is appended to the plurality of attributes.
15 . The system of claim 1 , wherein the first copy is written into the third database data chunk-by-data chunk where each such data chunk is 1 megabyte or less in size.
16 . The system of claim 1 , wherein the computing terminal is positioned within a pharmacy.
17 . The system of claim 1 , wherein the first data format is a National Council for Prescription Drug Programs (NCPDP) data format.
18 . The system of claim 1 , wherein the second data format is a structured data format or a delimited data format.
19 . The system of claim 1 , wherein the result is binary indicates an approval or a denial of the request.
20 . The system of claim 1 , wherein the at least one rule conditions or requires at least one metadata element sourced from the second copy, wherein the second data format is the metadata format.
21 . The system of claim 1 , wherein the first copy in the metadata format is written into the third database data chunk-by-data chunk based on a size of a data chunk.
22 . The system of claim 1 , wherein the result is a binary result, wherein the plurality of indicators is a plurality of binary indicators.
23 . A method, comprising:
operating a system, comprising:
a serverless environment hosting a plurality of compute nodes distributively running a parsing logic, an execution logic, and a rule engine, wherein the rule engine accesses a plurality of rules;
a persistent storage containing a first serverless database, a second serverless database, and a third serverless database, wherein the first serverless database stores a plurality of executable files enabling a plurality of statistical models, wherein the second serverless database stores a plurality of configurations profiles, wherein the third serverless database stores a plurality of historical metadata enabling a plurality of historical trends; and
a computing terminal hosting an application program such that the application program generates a request in a first data format and sends the request in the first data format to the parsing logic, wherein the request has a plurality of attributes, wherein without real-time provisioning the plurality of compute nodes to handle the request during a time window:
(a) the parsing logic receives the request in the first data format thereby initiating the time window, converts the request into a second data format, and sends the request in the second data format to the execution logic,
(b) the execution logic receives the request in the second data format, queries the second serverless database to retrieve a subset of the plurality of configuration profiles based on selecting a subset of the plurality of attributes, queries the first serverless database to retrieve a subset of the plurality of executable files based on the subset of the plurality of configuration profiles, executes the subset of the plurality of executable files in parallel to respectively generate in parallel a plurality of outputs based on selecting a subset of the plurality of statistical models such that the plurality of outputs contains a plurality of indicators and a plurality of context identifiers identifying a subset of the plurality of historical trends associated with a plurality of requests submitted to the execution logic before the request, generate a first copy of the plurality of outputs in a metadata format, write the first copy in the metadata format into the third database data chunk-by-data chunk such that the first copy in the metadata format is stored in the plurality of historical metadata thereby updating the plurality of historical trends, generate a second copy of the plurality of outputs in the second data format, send the second copy in the second data format to the rule engine,
(c) the rule engine receives the second copy in the second data format, applies at least one rule of the plurality of rules to the second copy such that a result is generated in the second data format and sent to the parsing logic, and
(d) parsing logic receives the result in the second data format, converts the result from the second data format to the first data format and sends the result in the first data format to the application program thereby terminating the time window.