International Business Machines Corporation

David R Vyvyan

Graham Anthony Bent

Patrick Dantressangle

Robert Neugebauer

Dmitri Abrashkevich

Ioanna Maria Attarian

Thomas J C Berman

Michael Cobbett

Andrew David James Banks

Mark Andrew Wynn-MacKenzie

Seraphin Bernard Calo

Fan Ye

Bernard Z Kufluk

Christopher R Gibson

Philip Noel

Nigel Lewis Jones

Lu Su

Harmanraj Singh Wadhwa

Bob Neugebauer

Aggregation in a computing system can include receiving, at a service node of the computing system, a first query specifying aggregation and translating the first query into a second query having a first canonical format and specifying the aggregation. The method can include forwarding the second query to a first subset of a plurality of endpoint nodes and translating, at each endpoint node of the first subset, the second query into a third query having a format executable by a data source connected to the endpoint node. The third query can specify a level of the aggregation to be performed by the data source determined based upon a processing capability of the data source. The endpoint nodes can initiate execution of the third query by the data sources and provide an aggregated result including a result from the data source(s) to the service node.

Aggregation in dynamic and distributed computing systems

A method includes: a first node and a second node in a distributed computing system each creating a respective partial join filter; the first node and the second node each transmitting its respective partial join filter to a third node in the distributed computing system; the third node creating a final join filter by combining the respective partial join filters of the first node and the second node; the third node retrieving target data from a data source of the third node by applying the final join filter to the data source of the third node; and the third node transmitting the retrieved target data to a controlling node in the distributed computing system.

Distributed join filters in a dynamic distributed data processing service

Table discovery in a distributed and dynamic computing system can include detecting an event of a type indicating an update to a data source of a selected node of a plurality of nodes of a computing system, in response to the event, generating, within the selected node, a first collaborative trigger having a query identifier correlated with the type of the event, and distributing the first collaborative trigger through other ones of the plurality of the nodes of the computing system so that the first collaborative trigger reaches an actor of the computing system. Table discovery further can include, in response to receiving the first collaborative trigger within the actor, generating, within the actor, a query for table definition information. The query may be implemented as an aggregation query and may be distributed through the plurality of the nodes.

Table discovery in distributed and dynamic computing systems

Using collaborative triggers within a distributed computing system can include, in response to detecting a first event of a selected type within a first node of the computing system, generating a first collaborative trigger having a query identifier correlated with the selected type of the event and distributing the first collaborative trigger through other nodes of the computing system so that the first collaborative trigger reaches an actor of the computing system. The computer-implemented method can include suppressing, within each node that distributes the first collaborative trigger, each other collaborative trigger having a same query identifier as the first collaborative trigger until a response to the first collaborative trigger is received from the actor.

Collaborative triggers in distributed and dynamic computing systems

Determining common table definitions in distributed databases includes receiving a query at a node in a distributed database. The query targets a logical table. Upon determining the logical table is locally defined, one of the local definition and a network definition is selected to use. The network definition is independently established for shared logical table names. The selected definition is used and propagated with the query through the distributed database. Upon determining the logical table is not locally defined, an initial query is issued to other nodes in network to find the network definition for the logical table, the network definition is stored in memory, and the selected definition is propagated through the distributed database with the query.

Determining common table definitions in distributed databases

Method and system are provided for assessing response routes in a network of connected nodes wherein a querying node propagates a query through multiple routes of intermediate nodes to a data source node and the data source node returns data to the querying node. The method at a data source node may include: receiving a query instance including metadata with route information of the route of intermediate nodes in the network that the query instance has travelled through from a querying node to the data source node; comparing the route information to a route policy; and responding to a query instance which meets the route policy by returning the query response via the route of intermediate nodes of the query instance.

Assessing response routes in a network

A query is propagated from a querying node to a destination node in a network of connected nodes, where at least one node is designated as a delay node. A query is transmitted from the querying node to one or more connected nodes in the network of connected nodes. One or more nodes, from the one or more connected nodes, are designated as delay nodes, wherein delay nodes apply delays to the query before retransmitting the query to other nodes from the one or more connected nodes. Upon receiving the query, each node, from the one or more connected nodes, retransmits the query to another one or more nodes from the one or more connected nodes. The query is repeatedly received and retransmitted by the connected nodes until the query is received at the destination node.

Propagating a query in a network

In at least one embodiment, a method and a system include a node potentially having information responsive to an information request distributed into, for example, a federated coalition network where the node receives at least one information request packet, conducts a search of information at the node to determine if requested information is present, when the requested information is present, then the node sends an acknowledgement to a requesting node, linear network codes the requested information into m packets where m is greater than or equal to k, which is the number of packets needed to be received by the requesting node to reconstruct the requested information, selects multiple paths between the node and the requesting node such that no third party will see more than k&#x2212;1 different packets, and transmits the m packets distributed over the selected paths. In a further embodiment, the method and system include operation of the network to have the capability to provide responsive information in a protected way.

Privacy preserving query method and system for use in federated coalition networks

A database query is received that includes a logical query indicator at a distributed database node within a distributed network of databases. The logical query indicator includes at least one physical database query and at least one database node identifier that allows at least one distributed database node to identify at least one physical database to execute the physical database query against. It is determined that the at least one database node identifier matches a local node identifier. The at least one physical database query is executed against at least one local physical database table. A local query response is formed including data retrieved from the at least one local physical database table. The database query is responded to with at least the local query response.

Node-level sub-queries in distributed databases

Determining common table definitions in distributed databases includes independently establishing, for a node in a distributed database, a common network definition for shared logical table names. Establishing the common network definition includes providing local definitions for at least one logical table, and querying at least one node in a network of nodes to determine any existing network definitions for the logical table. If there is no existing network definition, the method includes adding the local definition as a new network definition. If there is an existing network definition, the method includes updating the existing network definition with any extra columns of local definition, and applying an updated network definition to other nodes in the network holding a record of the network definition.

A database query is received that includes a logical query indicator at a distributed database node within a distributed network of databases. The logical query indicator includes at least one physical database query and at least one database node identifier that allows at least one distributed database node to identify at least one physical database to execute the physical database query against. It is determined that the at least one database node identifier matches a local node identifier. The at least one physical database query is executed against at least one local physical database table. A local query response is formed including data retrieved from the at least one local physical database table. The database query is responded to with at least the local query response. This abstract is not to be considered limiting, since other embodiments may deviate from the features described in this abstract.

A database query comprising a logical table definition and at least one route tracing indicator is forwarded to at least one distributed database node within a distributed network of databases. At least one query response is received including distributed route tracing information associated with each distributed database node that processed the database query along a path associated with the at least one query response. The received distributed route tracing information is processed to characterize query route propagation within the distributed network of databases. This abstract is not to be considered limiting, since other embodiments may deviate from the features described in this abstract.

Growing community of inventors

Hampshire, United Kingdom

David R Vyvyan