| CPC G16H 10/60 (2018.01) [G06F 16/258 (2019.01); G06F 16/27 (2019.01)] | 21 Claims |

|
1. A computer-implemented method, comprising:
receiving raw data with metadata indicating a set of originating sources of the raw data and a set of data types associated respectively with the set of originating sources;
referencing a plurality of rules that govern (a) filtering and (b) delivery of the raw data and that define subscriptions that comprise a plurality of data sink definitions, wherein the plurality of data sink definitions includes:
a first data sink definition specifying a first target data format for a first processing node in a plurality of processing nodes, the first processing node comprising a first memory and first set of processors configured to perform low-latency processing on the raw data and to implement a first computing solution,
a second data sink definition specifying a second target data format for a second processing node in the plurality of processing nodes, the second processing node comprising a second memory and second set of processors configured to perform low-latency processing on the raw data and to implement a second computing solution;
determining, based on at least the plurality of rules and the metadata, that a first portion of the raw data is subscribed to by the first processing node and that a second portion of the raw data is subscribed to by the second processing node;
determining, based on at least the plurality of rules and the metadata, that a third portion of the raw data is not subscribed to by any of the plurality of processing nodes;
excluding the third portion of the raw data from low-latency processing by the first processing node and the second processing node;
based on at least the first data sink definition and the second data sink definition, generating transformed data by modifying the first portion of the raw data into the first target data format and modifying the second portion of the raw data into the second target data format; and
communicating the transformed data to at least one processing node selected from the first processing node and the second processing node such that the at least one processing node can perform at least one computing solution selected from the first computing solution and the second computing solution.
|