摘要:
A compare, swap and store facility is provided that does not require external serialization. A compare and swap operation is performed using an interlocked update operation. If the comparison indicates equality, a store operation is performed. The compare, swap and store operations are performed as a single unit of operation.
摘要:
Performance counters are provided for virtualized network interfaces of communications networks, while minimizing the use of hardware resources. A virtualized network interface includes physical resources, as well as logical resources. Dedicated performance counters are provided for the physical resources of the virtualized network interface, as well as for logical partitions coupled to that interface, while non-dedicated performance counters are provided for the logical resources. This enables the provision of performance counters for virtualized network interfaces, while minimizing hardware resources consumed by those interfaces.
摘要:
A method and apparatus of operating a central processing unit (CPU) including a plurality of processors, is provided and includes collecting real-time statistics relating to the processors during dispatching activities, identifying give-help processors from the real-time statistics when the real-time statistics indicate that one or more of the nodes is overworked, and implementing help to be provided by the give-help processor to relieve the overworked node of a portion of the work to be distributed thereto.
摘要:
An information processing system is provided which includes a plurality of system resources, and an event queue having a maximum number of entries. An event recording mechanism of the information processing system is operable to make entries regarding events in the event queue, wherein the entries are limited to a predetermined number of active entries in the event queue per each type of event per each of the system resources. In a particular embodiment, the number of entries per each type of event for each of the system resources is limited to one.
摘要:
A computer-implemented method for distributing a plurality of tasks over a plurality of processing nodes in a processor network includes the following steps: calculating a task process consumption value for the tasks; calculating a measured node processor consumption value for the nodes; calculating a target node processor consumption value for the nodes, the target node processor consumption value indicating optimal node processor consumption; calculating a load index value as a difference between the calculated node processor consumption value for a node i and the target node processor consumption value for node i; and distributing the tasks among the nodes to balance a processor workload among the nodes, according to the calculated load index value, such that the calculated load index value of each node is substantially zero. The method further embodies a multi-dimensional balancing matrix, each dimension of the matrix representing a node corresponding to a different processor type and each cell representing tasks assigned to multiple nodes.
摘要:
An information processing system is provided which includes a plurality of system resources, and an event queue having a maximum number of entries. An event recording mechanism of the information processing system is operable to make entries regarding events in the event queue, wherein the entries are limited to a predetermined number of active entries in the event queue per each type of event per each of the system resources. In a particular embodiment, the number of entries per each type of event for each of the system resources is limited to one.
摘要:
An apparatus, method and program product for sending data to or receiving data from one or more I/O devices in an I/O operation with a main storage controlled by a processor in a data processing system. The apparatus includes a time-of-day (TOD) register for containing a TOD value, a clock for containing a current TOD value, and a summary register having a first condition when any one of said devices requests an I/O operation and a second condition when no devices have an outstanding I/O request, each device having an outstanding I/O request sets the summary register to its first condition only when the summary register is in its second condition, and further places the current TOD value in the TOD register. A checking program determines if a specified time delay has been exceeded between the value in said TOD register and the current TOD for each requested I/O operation. The checking program drives an interrupt to the processor when the specified time delay has been exceeded.
摘要:
A method of providing direct data processing access in a network computing system environment. The system environment has a main storage which can be connected to one or more application servers and is in processing communication with an interface element. The interface element has at least one adapter and can be connected to one or more application user(s). One or more queues are established in the main storage that can handle data without causing interrupts in the running programs. Incoming data is received using the adapter and as data is received or modified, the status of the network computing system will be updated to reflect the new data or change. Data is then processed in the main storage after interrogating the multiple existing queues in the main storage and forwarding them in turn to their appropriate destination or application server after a determination has been made by interrogating these queues.
摘要:
Distributed computing environment (DCE) remote procedure calls (RPCs) are integrated with an advisory work load manager (WLM) to provide a way to intelligently dispatch RPC requests among the available application server processes. The routing decisions are made dynamically (for each RPC) based on interactions between the location broker and an advisory work load manager. Furthermore, when the system contains multiple coupled processors (tightly coupled within a single frame, or loosely coupled within a computing complex, a local area network (LAN) configuration, a distributed computing environment (DCE) cell, etc.), the invention extends to balance the processing of RPC requests and the associated client sessions across the coupled systems. Once a session is assigned to a given process, the invention also supports performance monitoring and reporting, dynamic system resource allocation for the RPC requests, and potentially any other benefits that may be available through the specific work load manager (WLM).
摘要:
A method establishes a transport layer connection between a first system and a second system. The establishment of the transport layer connection includes identifying a remote direct memory access (RDMA) connection between the first system and the second system. After establishing to transport layer connection, the first and second systems exchange data using the RDMA connection identified in establishing the transport layer connection.