Abstract: A high performance computer system has a number of compute nodes interconnected by an inter-node communication network. Each compute node has a local packetized interconnect coupled to the inter-node communication network by an interface. Data packets on the local packetized interconnect of a sending node may be delivered to a destination in a receiving node by addressing them to addresses associated with the network interface and tunneling the packets through the inter-node communication network to the receiving node.

Abstract: Compute nodes in a high performance computer system are interconnected by an inter-node communication network. Each compute node has a network interface coupled directly to a CPU by a dedicated full-duplex packetized interconnect. Data may be exchanged between compute nodes using eager or rendezvous protocols. The network interfaces may include facilities to manage data transfer between computer nodes.

Abstract: An apparatus and method for replacing a CPU cooling system, for example, a fan, while a computer continues to operate has a mechanism for maintenance personnel to signal that cooling fan replacement is about to begin. In response to the signal, a control mechanism causes the CPU to shift into an operating mode in which heat generation is reduced. In some cases, a clock frequency of the affected CPU and/or a duty cycle at which the CPU operates is reduced. After the fan has been replaced the control mechanism returns the CPU to a normal operational mode, for example, by increasing the clock frequency or duty cycle to a normal level. During the fan replacement procedure, the CPU and its software continues to run at lower than normal performance.

Abstract: A network topology links nodes with Cartesian and diagonal links. The diagonal links may be uniformly-extending iso-delta diagonal links of different lengths. The network topology may be used to provide data communication between processors or groups of processors in a high-performance computer system.