Abstract: Systems and methods are described herein that include flow control mechanisms that provide a receiving device with the ability to reclaim buffers that have been previously advertised to a sending device. Data structures and communication methods are described that facilitate the communication of flow control messages between sending and receiving devices that allow an advertised window to be reduced, and buffers to be released, by a sending device in response to a flow control message from the receiving device.

Abstract: Systems and methods are described herein that include flow control mechanisms that provide a receiving device with the ability to reclaim buffers that have been previously advertised to a sending device. Data structures and communication methods are described that facilitate the communication of flow control messages between sending and receiving devices that allow an advertised window to be reduced, and buffers to be released, by a sending device in response to a flow control message from the receiving device.

Abstract: Techniques are provided to allow users to define a global service that is offered across multiple replicated databases. Database clients connect to and use a global service just as they do with regular services on a single database today. Upon receiving a connection request, a collection of components, referred to collectively as the Global Data Service framework (GDS framework), automatically chooses the best database server instances to which to connect a client. Once those connections have been established, the clients determine which database server instance, of those database server instances to which they are connected, to send requests to based, at least in part, on advisory messages sent to the clients by the GDS framework.

Abstract: Techniques are provided to allow users to define a global service that is offered across multiple replicated databases. Database clients connect to and use a global service just as they do with regular services on a single database today. Upon receiving a connection request, a collection of components, referred to collectively as the Global Data Service framework (GDS framework), automatically chooses the best database server instances to which to connect a client. Once those connections have been established, the clients determine which database server instance, of those database server instances to which they are connected, to send requests to based, at least in part, on advisory messages sent to the clients by the GDS framework.

Abstract: The techniques are provided automatically detecting when performing pre-transmission compression on data to be sent over a network connection will be preferable to sending the data uncompressed, and for automatically performing pre-transmission compression only when doing so is determined to be beneficial. The techniques involve performing compression on sample data. The time it takes to perform the compression on the sample data, along with the reduction in size achieved by the compression, are factors used to automatically determine whether compressing data before sending the data over a network connection will achieve better results that sending the data uncompressed. In some embodiments, multiple compression algorithm/compression level combinations are tested at different points in the transmission. At each point, the sending device may switch to the algorithm/level combination that is producing the best results on the current sample.

Abstract: Techniques are described herein for propagating continuously generated data from a seed to a plurality of peers without requiring the seed to send the new data to all peers, and without establishing static sender/receiver relationships between the peers. The data generated by the seed is divided into a number of “windows”, where each window is divided into a number of “pieces”. The seed sends the data piece-by-piece to different peers. The peer to which the seed sends any given piece may be selected at random. The peers exchange pieces until all pieces of a window are obtained by each peer. When a peer receives all pieces of a window, the pieces are applied at the peer, and the process repeats for the next window.

Abstract: Techniques are provided to allow users to define a global service that is offered across multiple replicated databases. Database clients connect to and use a global service just as they do with regular services on a single database today. Upon receiving a connection request, a collection of components, referred to collectively as the Global Data Service framework (GDS framework), automatically chooses the best database server instances to which to connect a client. Once those connections have been established, the clients determine which database server instance, of those database server instances to which they are connected, to send requests to based, at least in part, on advisory messages sent to the clients by the GDS framework.

Abstract: Techniques and systems that allow receiving a data stream and a location value. The location value, in one embodiment, is indicative of a location in the data stream at which the data stream has been aborted. This value may be determined by a sending entity and sent to a receiving entity. In various embodiments, the receiving entity may compute the remaining amount of data to be received in the data stream, and then receive that amount of data. In some embodiments, a checkpoint value may be used in conjunction with the location value to indicate an abort location for a data stream. A checkpoint value may correspond to an amount of data between successive checkpoints in the data stream. In some embodiments, upon aborting a data stream, a receiving entity receives data until a next checkpoint in the data stream.

Abstract: The techniques are provided automatically detecting when performing pre-transmission compression on data to be sent over a network connection will be preferable to sending the data uncompressed, and for automatically performing pre-transmission compression only when doing so is determined to be beneficial. The techniques involve performing compression on sample data. The time it takes to perform the compression on the sample data, along with the reduction in size achieved by the compression, are factors used to automatically determine whether compressing data before sending the data over a network connection will achieve better results that sending the data uncompressed. In some embodiments, multiple compression algorithm/compression level combinations are tested at different points in the transmission. At each point, the sending device may switch to the algorithm/level combination that is producing the best results on the current sample.

Abstract: Techniques are described herein for propagating continuously generated data from a seed to a plurality of peers without requiring the seed to send the new data to all peers, and without establishing static sender/receiver relationships between the peers. The data generated by the seed is divided into a number of “windows”, where each window is divided into a number of “pieces”. The seed sends the data piece-by-piece to different peers. The peer to which the seed sends any given piece may be selected at random. The peers exchange pieces until all pieces of a window are obtained by each peer. When a peer receives all pieces of a window, the pieces are applied at the peer, and the process repeats for the next window.

Abstract: Systems, methods, and other embodiments associated with set transparent network communication are described. In one embodiment, in response to receiving a listen signal from a listen requestor, where the listen signal contains a handle identifying a set of interfaces from which the listen requestor is configured to receive incoming network communications: a set of listening endpoints are opened, where a member of the set of listening endpoints is associated with a member of the set of interfaces. The handle is modified to produce a modified handle that identifies the set of listening endpoints. The modified handle facilitates set transparent communication between the listen requester and a member of the set of listening endpoints. The modified handle is provided to the listen requester.

Abstract: Systems, methods, and other embodiments associated with set transparent network communication are described. In one embodiment, in response to receiving a listen signal from a listen requestor, where the listen signal contains a handle identifying a set of interfaces from which the listen requestor is configured to receive incoming network communications: a set of listening endpoints are opened, where a member of the set of listening endpoints is associated with a member of the set of interfaces. The handle is modified to produce a modified handle that identifies the set of listening endpoints. The modified handle facilitates set transparent communication between the listen requester and a member of the set of listening endpoints. The modified handle is provided to the listen requester.

Abstract: An improved approach is described for sharing resources such as sessions and connections in database systems. Conduits and templates in shared memory can be employed to facilitate the transfer of network sessions. This provides a very efficient way to share processing entities across multiple users or clients.