Abstract: A method and apparatus for distributing multiple interrupts among multiple processors is disclosed. According to one embodiment, an interrupt daemon monitors the interrupt load among the processors that results from an initial mapping of the interrupts to the processors. The interrupt daemon determines whether there is a sufficient imbalance of the interrupts among the processors. If so, the interrupt daemon triggers a reassignment routine that generates a new mapping of the interrupts among the processors, and if not, the interrupt daemon goes to sleep for a specified time period. If the new mapping produces a sufficient improvement in the distribution of interrupts among the processors, based on the same criteria used to detect the imbalance, the new mapping is used by the central hub for subsequent distribution of interrupts to the processors. However, if the new mapping does not provide a sufficient improvement, the original mapping continues to be used.

Abstract: One embodiment of the present invention provides a system that buffers data inside of a byte-stream protocol at a transport layer on a client, which is receiving the data from an external source. The system operates by receiving a configuration parameter at the transport layer from an application executing on the client, wherein the configuration parameter specifies a condition upon which data buffered at the transport layer is to be sent to the application. The system then buffers data destined for the application at the transport layer on the client, and sends the buffered data to the application when the condition specified by the configuration parameter is met. Buffering data at the transport layer on the client facilitates more efficient processing of system calls.

Abstract: One embodiment of the present invention provides a plug-in architecture for a network stack in an operating system. The network stack includes a set of functions configured to modify a set of parameters that are likely to change based on the network environment. The architecture includes a plug-in framework within the network stack that allows the set of functions to be dynamically changed in order to change the TCP behavior of the network stack to suit the network environment.

Abstract: One embodiment of the present invention provides a system that dynamically changes the TCP behavior of a network connection. First, the system receives a request to change the TCP behavior for a network connection that allows communication between a first computer system and a second computer system. In response, the system changes a function associated with the TCP behavior of the network connection to a new function that provides TCP behavior better-tuned to the needs and environment of the network connection.

Abstract: One embodiment of the present invention provides a system for sending data to a remote host using a socket. During operation the system receives a request from an application to write data to the socket, wherein the data is stored in a source memory buffer in user memory. Next, the system initiates a DMA (Direct Memory Access) transfer to transfer the data from the source memory buffer to a target memory buffer in a TCP (Transmission Control Protocol) Offload Engine. The system then returns control to the application without waiting for the TCP Offload Engine to send the data to the remote host.

Abstract: A system and method of operating a client network computer in a disconnected mode. A client computer system includes a client storage device, a processor, a network interface, a failover server and a software manager. The processor is configured to execute software instructions stored in the client storage device. The network interface is configured to connect the client computer system to a remote network server unit. A failover server implemented on the client computer system is configured to provide functionality similar to the remote network server unit by accessing a copy of a network database file stored on the client storage device. A software manager stored in the client storage device is configured to cause the client computer system to connect to the remote network server unit if the remote network server unit is available or to cause the client computer system to connect to the failover server if the remote network server unit is not available.

Abstract: One embodiment of the present invention provides a system that facilitates implementing an Internet Protocol (IP) link over an InfiniBand™ (IB) network. During operation, the system receives a Partition Key (P_Key) and a Queue Key (Q_Key) defined within the IB network. Next, the system associates the IP link with the combination of the P_Key and the Q_Key so that operations involving the IP link are mapped to a subset of the nodes in the IB network defined by the combination of the P_Key and the Q_Key.