Abstract: Dynamic data clock rate for data transfer to a display. The rate at which data is transferred from memory by a processor to a display is controlled such that data is transferred at or near one or more pixel refresh rates specified for the display while allowing at least one of the processor or memory to perform other tasks. A rate at which data is transferred from the processor to the display is reduced when the processor and/or memory is burdened with other tasks. When the processor and/or memory become more available as a result of reduction or elimination of other tasks, the rate at which data is transferred from the processor to the display is increased.

Abstract: A method for dual-processor communication. In one example embodiment, a method includes indicating to a master processor that data is available to transfer to the master processor; receiving a request for the data from the master processor over a storage-based channel; and sending the data to the master processor over the storage-based channel. In another embodiment a method includes determining that data is available for transfer from a slave processor; sending a request for the data to the slave processor over a storage-based channel; and receiving the data from the slave processor over the storage-based channel. In another embodiment a method includes detecting an event on a master processor and writing data corresponding to the event to a slave processor over a storage-based channel.

Abstract: A system and method for dynamically timing the sending of an acknowledgment packet across a network. In one example embodiment, such a method includes determining that a receive data window contains unacknowledged data, determining whether the time since receiving the earliest unacknowledged data packet in the receive data window is less than a pre-determined deferred acknowledgment time, and determining whether the portion of the receive data window with unacknowledged data is less than a dynamically-determined percentage. If the time since receiving the earliest unacknowledged data packet in the receive data window is less than the pre-determined deferred acknowledgment time and the portion of the receive data window with unacknowledged data is less than the dynamically-determined percentage, the method further includes deferring the sending of an acknowledgment packet.

Abstract: Caching meta information including file relationships at a local processor. A local processor requests files and directories from the remote file system by sending a network identifier used by the remote file system. At the local processor, a network identifier is used to access a remote upper level directory at the remote file system. Meta information is obtained from the remote file system about files or directories hierarchically below the upper level directory. The meta information about the files or directories hierarchically below the upper level directory is cached along with relationship information about the files or directories without storing network identifiers for the files or directories. A network identifier is cached or known for a directory hierarchically above the files or directories that can be used to obtain the network identifiers for desired files or directories from the remote file system.

Abstract: Accessing data from large, unknown size, or streaming files. A method includes acts for accessing files. A request is received to read at a virtual address corresponding to an identifier for a file or directory. The virtual address is read to read a shared address range from the virtual address. The shared address range can be used for a number of files or directories. One of the files or directories may be selected for use at a given time based on the identifier for the file or directory. The shared address range is returned in response to the request. File or directory content data is accessed by means of a read request, received for an address in the shared address range, wherein the shared address range comprises a finite address range, and wherein the shared address range is circular such that if a file or directory exceeds the size of the finite address range, portions exceeding the finite address range can be addressed starting at the beginning of the shared address range.

Abstract: A system for accessing data across a network. The system includes a first processor configured to consume data. A second processor is connected to the first processor via a physical media storage interface. As such, the first processor interfaces with the second processor as if the second processor were a physical media storage device. The second processor is connectable via a network interface to a physical media storage device so as to be able to provide data from the physical media storage device to the first processor. The system includes a computer readable medium including computer executable instructions configured to compensate for at least one of time latencies or network connection problems caused by the network interface between the second processor and the physical media storage device.