Method for performing multi-caching on data sources of same type and different types by using cluster-based processing system and device using the same

A cluster management device for performing multi-caching on data sources of different types by using a cluster-based processing system connected to an external data source, the cluster-based processing system including a master node and a worker node, the cluster management device comprising:

• at least one memory that stores instructions;
• and
• at least one processor configured to execute the instructions to perform or support another device to perform a process comprising:
• acquiring a user query that comprises a plurality of query elements;
• determining whether or not a complete response to the plurality of query elements pre-exists as joined data within a first cache of either the master node or the worker node;
• upon determining that a complete response to the user query does not pre-exist as joined data within the first cache of either the master node or the worker node, parsing the user query into the plurality of query elements;
• determining whether or not individual responses to each the plurality of query elements exists as individual data elements within a second cache of either the master node or the worker node, thereby producing a determination result;
• the determination result indicating that:
• first joinable data corresponding to a first subset of the plurality of query elements exists as individual data elements within the second cache of either the master node or the worker node,
• first unjoined data corresponding to a second subset of the plurality of query elements exists as an individual data element within the second cache of either the master node or the worker node, and
• second unjoined data corresponding to the second subset of the plurality of query elements does not exist within the second cache of either the master node or the worker node;
• based on determining that the second unjoined data corresponding to the second subset of the plurality of query elements does not exist within the second cache of either the master node or the worker node, determining that the second unjoined data corresponding to the second subset of the plurality of query elements does exist as an individual data element within the external data source;
• performing a series of JOIN operations to create the complete response to the plurality of query elements, the JOIN operations comprising:
• joining the first joinable data to form first joined data,
• joining the first unjoined data within the second cache of either the master node or the worker node and second unjoined data within the external data source to form second joined data, and
• joining the first joined data with the second joined data to produce the complete response to the plurality of query elements;
• and
• reporting the complete response to the plurality of query elements as a reply to the user query.