Systems and methods for low latency aggregated data provision

A computer-implemented method for low-latency aggregated-data provision, the method comprising:

• receiving, via a materialization service, data from one or more sources related to an item associated with a webpage hosted by a web server, the data including:
• first event data from a first data store, the first event data sourced from a real-time feed, and
• second event data from a second data store, the second event data sourced from stored data;
• generating synchronized data at the materialization service, based on the first event data and second event data;
• receiving a request from the web server for information related to the item;
• identifying data related to the first event data and the second event data;
• generating, based on the identified event related data and a user segment, a data structure structured to initialize an aggregation process, the data structure comprising a directed acyclic graph and wherein generating the data structure comprises generating an earlier node of the directed acyclic graph based on a later node of the directed acyclic graph such that the data structure is generated backwards from the later node, wherein the later node corresponds to a desired result associated with the request;
• aggregating the synchronized data by following a path through nodes of the directed acyclic graph of the data structure, wherein the path through the nodes of the directed acyclic graph is dynamically identified to optimize aggregation of the synchronized data;
• and
• forwarding the aggregated synchronized data to the web server in response to the request.

A computer-implemented system for low-latency aggregated-data provision, the system comprising:

• at least one memory storing instructions;
• and
• at least one processor configured to execute the instructions to perform operations comprising:
• receiving, via a materialization service, data from multiple sources related to an item associated with a webpage hosted by a web server, the data including:
• first event data from a first data store comprising a key-value store, the first event data sourced from a real-time feed, and
• second event data from a second data store comprising a clustered database, the second event data sourced from stored data comprising primitive data;
• generating synchronized data at the materialization service, based on the first event data, and the second event data, and a timestamp of the first event data or the second event data;
• receiving a request from the web server for information related to the item;
• determining a user segment with the request based on at least one of a location, a user history, or a promotion condition;
• identifying data related to the first event data and the second event data;
• generating, based on the identified event related data and the user segment, a data structure structured to initialize an aggregation process, wherein the data structure comprises a directed acyclic graph and wherein generating the data structure comprises generating an earlier node of the directed acyclic graph based on a later node of the directed acyclic graph such that the data structure is generated backwards from the later node, wherein the later node corresponds to a desired result associated with the request;
• aggregating the synchronized data by following a path through nodes of the directed acyclic graph of the data structure, wherein the path through the nodes of the directed acyclic graph is dynamically identified to optimize aggregation of the synchronized data;
• and
• forwarding the aggregated synchronized data to the web server in response to the request.