Abstract: Apparatus and method that schedules movement of packets within network devices, such as network processors, includes a calendar using a segmented hierarchical routine to identify the next packet to be moved from one of a plurality of flow queues.

Abstract: A method is disclosed for discovering servers, spawning collector threads to collect information from servers, and reporting such information. The method may determine a number of servers communicatively coupled to a network. For each server, a collector thread may be spawned to collect information regarding the server by sending requests to the server and receiving responses from the server. The collector threads may be spawned by and run on a computing device other than the number of servers, such that no computer-executable code is installed on the servers for collecting the information. Upon completion of the collector thread for each server, the information regarding the server as collected may be stored to a database by one or more writer threads.

Abstract: Controlling the merging of nodes into a single-partition merged system using the nodes' service processors is disclosed. The system runs a single instance of an operating system. A user communicates with a service processor of a predetermined boot node to enter partition configuration information for the system without using a dedicated console. The partition configuration information is stored by the service processor of the boot node, and transmitted from the service processor of the boot node to the service processors of predetermined secondary nodes. The service processor of each secondary node then stores the partition configuration information. At reset, power-up, or power-down of any secondary node, the reset, power,-up or power-down is conveyed to the service processor of the boot node, which resets, starts up, or shuts down the system, including transmitting commands to the service processors of the secondary nodes to individually reset, start up, or shut down.

Abstract: Memory allocation requests are satisfied from a memory pool or from lookaside lists, based on the memory size requested to be allocated. A method, for each of a number of future differently sized memory allocations, determines which memory allocations are to be satisfied from a lookaside list, and which memory allocations are to be satisfied from a memory pool. For each memory allocation to be satisfied from a lookaside list, a corresponding lookaside list for the allocation is initialized. A table is constructed that has a number of entries corresponding to all the differently sized memory allocations. Each entry corresponding to a memory allocation to be satisfied from a lookaside list points to the lookaside list for that memory allocation.

Abstract: An AC magnetic erase head having improved AC erase characteristics, the magnetic head being formed of two half core elements of a magnetic material, preferably ferrite, defining an irregular edged or jagged-edged gap, and including a non-magnetic substance, preferably glass, within the gap. The magnetic head is preferably produced by thermally etching ferrite magnetic material forming the half core elements by means of relatively prolonged heat soak at an elevated temperature.