Apparatus and method for determining key performance indicators

A data processing apparatus deployed at a cloud network, the apparatus comprising:

• a plurality of data nodes;
• a receiver circuit coupled to the plurality of data nodes and configured to receive downtime records from a plurality of industrial machines, and to selectively route the downtime records to selected ones of the plurality of data nodes based upon a predetermined criteria;
• wherein each of the plurality of data nodes comprises:
• a control circuit;
• and
• a memory coupled to the control circuit, the memory storing selected one of the downtime records;
• an aggregation circuit coupled to the plurality of data nodes;
• wherein each control circuit is configured to, in parallel with the other control circuits, further populate the downtime records with other data related to the operation of the machines, to determine one or more time windows for each of the records, and to divide the downtime records according to the time windows, wherein the downtime records include a first data record covering events in a first overall time window, wherein the first data record is physically split at a first data node into a first sub-record and a second sub-record, the first sub-record corresponding to a first time window and the second sub-record corresponding to a second time window, wherein the first time window and the second time window collectively cover events in the first overall time window associated with the first data record;
• wherein the downtime records also include a second data record covering events in a second overall time window, wherein the second data record is physically split at a second data node into a third sub-record and a fourth sub-record, the third data sub-record corresponding to a third time window and the fourth sub-record corresponding to a fourth time window, wherein the third time window and the fourth time window collectively cover events in the second overall time window associated with the second data record, and wherein the splitting of the first data record at the first data node and the second data record at the second data node occurs in parallel;
• wherein a query is received and forwarded to the each of the plurality of data nodes, wherein the query relates to a performance indicator associated with one or more of the plurality of industrial machines;
• wherein in response to the query, each control circuit of each data node, in parallel with the other control circuits of others of the plurality of data nodes, selectively examines the downtime records in the memory at each data node according to the time windows, and forms an answer to the query;
• wherein each of the answers from the data nodes is aggregated by the aggregation circuit to form a final answer to the query.