Abstract: A method of parallel processing by networked devices, in which distributed data in a network is accessed in parallel to provide results. Commands and queries by a user station define a set of such distributed data. The networked devices access the distributed data in portions, individually process each portion, conduct peer-to-peer communication to aggregate results from processing each portion, and individually determine whether and when to report results. Making effective use of network assets (such as processing and storage at each node), making effective use of concurrency of network assets, limiting bandwidth use between network assets and user stations, and limiting use of assets (such as processing and storage) at user stations.

Abstract: A first managed machine of a plurality of managed machines arranged in a linear communication orbit based on a predefined order of respective machine identifiers of the managed machines scans for live unmanaged machines within a selected portion of the network that is associated with a first range of machine identifiers that includes identifiers between the respective identifiers of the first managed machine and a respective neighbor machine of the first managed machine, determines whether the remedial instruction specifies a respective remedial operation applicable to the one or more live unmanaged machines that have been detected within the selected portion of the network, and requests the one or more live unmanaged machines to execute the respective remedial operation specified by the remedial instruction.

Abstract: In one aspect, methods, system, and computer-readable media for monitoring unmanaged assets in a network having a plurality of managed machines include: at a first managed machine of the plurality of managed machines, wherein the plurality of managed machine are arranged in a linear communication orbit and have respective identifiers, and each managed machine is coupled to at least one respective neighbor by a corresponding local segment of the linear communication orbit: responding to a detection instruction for detecting unmanaged assets currently present in the network, by: scanning for live unmanaged machines within a selected portion of the network that is associated with a range of identifiers that includes identifiers between the respective identifiers of the first managed machine and a respective neighbor of the first managed machine; and generating a local report identifying one or more unmanaged machines that have been detected within the selected portion of the network.

Abstract: This application is directed to a distributed data processing method performed at a server system coupled to a linear communication orbit. The server system has a plurality of function modules. Each function module is configured to collect data related to a core function from the linear communication orbit. Each function module includes an internal client configured to adaptively perform a set of data processing operations according to a schema definition, including generating a data collection request for collecting raw data items, sending the data collection request through the linear communication orbit, collecting the requested raw data items from a set of machines via the linear communication orbit, and performing analysis on the collected raw data items. In some embodiments, a central data management module of the one or more servers is configured to provide the schema definition to and receive result data reported from the function modules.

Abstract: A respective node in a linear communication orbit receives an instruction packet through the linear communication orbit, where the instruction packet has been propagated from a starting node to the respective node through one or more upstream nodes along the linear communication orbit, and the instruction packet includes an instruction for establishing a direct duplex connection between the respective node and a respective server. In response to receiving the instruction packet, the respective node sends an outbound connection request to the respective server to establish the direct duplex connection. The respective node then uploads local data to the respective server through the direct duplex connection (e.g., in response to one or more queries, instructions, and requests received from the respective server through the direct duplex connection), where the respective server performs analysis on the local data received from the respective node through the direct duplex connection.

Abstract: In a distributed system, each of N machines receives a similarity search query through a linear communication orbit. The similarity search query includes token identifiers corresponding to tokens in a target document. Each machine, in response, identifies files that meet predefined similarity criteria with respect to the target document. Subsequent to receiving the similarity search query, the machine generates a first report, including a count of files stored at the machine that meet the predefined similarity criteria with respect to the target document, and/or information identifying a set of files that meet the predefined similarity criteria with respect to the target document; and sends the first report to a server through the linear communication orbit. The server produces a merged report presenting information with respect to files at a set of machines, including the N machines, that meet the predefined similarity criteria with respect to the target document.