Dynamic-ledger-enabled edge-device query processing
A method for processing a query for data stored in a distributed database includes receiving, at an edge device, the query for data stored in the distributed database from a query device. The method includes causing, by the edge device, the query to be stored on a dynamic ledger maintained by the distributed database. The method includes detecting, by the edge device, that summary data has been stored on the dynamic ledger. The method includes generating, by the edge device, an approximate response to the query based on the summary data stored on the dynamic ledger. The method includes transmitting, to the query device, the approximate response.
1 . A computer-implemented method for processing a query for data stored in a distributed database, the method comprising:
passing, by an edge device that is part of a distributed network of edge devices, the query for data stored in the distributed database;
detecting, by the edge device, summary data generated in response to the query, wherein the summary data is detected from one or more other edge devices in the distributed network;
generating, by the edge device, an approximate response to the query based on the summary data, wherein a probability distribution model is applied to generate the approximate response; and
transmitting the approximate response.
2 . The method of claim 1 , further comprising:
executing, by the edge device, a partial query against edge storage connected to the edge device to obtain partial query results;
generating, by the edge device, statistical information based on the partial query results,
wherein the statistical information includes at least one of: a mean, a standard deviation, a sum, or a count of the data,
wherein the probability distribution model is based on the partial query results, and
wherein the probability distribution model is configured to generate approximate responses to queries; and
determining, by the edge device, a statistical confidence associated with the probability distribution model based on at least one of:
a sample size of the partial query results,
a confidence interval threshold, or
a standard deviation of the partial query results.
3 . The method of claim 1 , wherein the transmitting the approximate response includes transmitting an associated confidence factor indicating an accuracy of the approximate response.
4 . The method of claim 1 , wherein the probability distribution model is trained on a training set of queries and responses of the distributed database.
5 . The method of claim 1 , wherein the probability distribution model includes at least one of a type or a location of the edge device.
6 . The method of claim 1 , wherein the probability distribution model includes a current condition of at least one of: a network environment of the edge device, a performance of the distributed database, a synchronization state of the distributed database, or a type of the distributed database.
7 . The method of claim 1 , wherein the probability distribution model includes at least one of: an intent of a query, a specified purpose of use of the edge device, or a priority level of a query.
8 . A computing system for processing a query for data stored in a distributed database, the system comprising:
an edge device that is part of a distributed network of edge devices; and
at least one processor that executes a set of computer-readable instructions, wherein, by executing the set of computer-readable instructions, the at least one processor collectively:
passes, by the edge device, the query for data stored in the distributed database;
detects, by the edge device, summary data generated in response to the query, wherein the summary data is detected from one or more other edge devices in the distributed network;
generates, by the edge device, an approximate response to the query based on the summary data, wherein a probability distribution model is applied to generate the approximate response; and
transmits the approximate response.
9 . The system of claim 8 , wherein the edge device is configured to:
receive an edge query language (EDQL) query specifying a distributed join operation;
identify required reference tables for executing the distributed join operation;
obtain the required reference tables from a dynamic ledger; and
execute the distributed join operation using the required reference tables.
10 . The system of claim 9 , wherein, by executing the set of computer-readable instructions, the at least one processor further:
analyzes historical query logs stored in edge storage;
predicts future queries based on identified patterns in at least the analyzed historical query logs;
pre-positions the required reference tables at edge devices based on the predicted future queries; and
optimizes query execution by leveraging edge shard algorithms and at least one of the pre-positioned reference tables to determine routing of queries to specific edge device partitions.
11 . The system of claim 8 , further comprising a platform that is integrated to link a set of systems for automation of management of supply activities within a value chain with a set of systems for automation of prediction of demand.
12 . The system of claim 8 , further comprising a platform that is integrated to link a set of systems for automation of management of supply activities within a value chain with a set of systems for automation of shaping of demand for entities handled by the value chain.
13 . The system of claim 8 , further comprising a platform that is integrated to enable centralized control of a set of value chain network entities.
14 . The system of claim 8 , further comprising a system that is integrated to enable at least one of: decentralization, cognition, or self-organization among a distributed set of value chain network entities.
15 . The system of claim 8 , further comprising a hybrid platform that is integrated, wherein the hybrid platform includes:
a platform that enables centralized control of a set of value chain network entities; and
a system that enables at least one of: decentralization, cognition, or self-organization among a distributed subset of the set of value chain network entities.
16 . A computer-implemented method for processing a query for data stored in a distributed database, the method comprising:
passing, by an edge device that is part of a distributed network of edge devices, the query for data stored in the distributed database;
detecting, by the edge device, summary data generated in response to the query, wherein the summary data is detected from a threshold of one or more edge devices that were previously used to respond to the query;
generating, by the edge device, an approximate response to the query based on the summary data; and
transmitting the approximate response.
17 . The method of claim 16 , wherein the approximate response is generated using a probability distribution model.
18 . The method of claim 17 , wherein the probability distribution model is trained on a training set of queries and responses of the distributed database.
19 . The method of claim 17 , wherein the probability distribution model includes at least one of a type or a location of the edge device.
20 . The method of claim 19 , wherein the probability distribution model includes a current condition of at least one of: a network environment of the edge device, a performance of the distributed database, a synchronization state of the distributed database, or a type of the distributed database.
21 . The method of claim 16 , wherein the summary data is stored on at least one of: a distributed ledger, a branch ledger, or a dynamic ledger.
22 . The method of claim 16 , wherein the generating, by the edge device, the approximate response to the query includes:
generating, by the edge device, a query plan for executing a complex query expression by:
identifying data components to be retrieved from edge storage, and
identifying components to be approximated using one or more probability distribution models; and
executing, by the edge device, the query plan by:
retrieving first data components from edge storage,
generating at least one approximation for second data components using the one or more probability distribution models,
determining one or more statistical confidence levels for the at least one approximation based on sample sizes and confidence interval thresholds;
processing, using a self-organizing neural network at the edge device, unlabeled sensor data to identify data patterns and sources; and
generating a query response combining the first data components and approximations when the one or more statistical confidence levels exceed defined thresholds.
23 . The method of claim 22 , wherein the transmitting the approximate response includes transmitting the one or more statistical confidence levels indicating one or more accuracy levels.