3185 results about "Multi processor" patented technology

A level of abstraction is created between a set of physical processors and a set of virtual multiprocessors to form a virtualized data center. This virtualized data center comprises a set of virtual, isolated systems separated by a boundary referred as a partition. Each of these systems appears as a unique, independent virtual multiprocessor computer capable of running a traditional operating system and its applications. In one embodiment, the system implements this multi-layered abstraction via a group of microkernels, each of which communicates with one or more peer microkernel over a high-speed, low-latency interconnect and forms a distributed virtual machine monitor. Functionally, a virtual data center is provided, including the ability to take a collection of servers and execute a collection of business applications over a compute fabric comprising commodity processors coupled by an interconnect. Processor, memory and I / O are virtualized across this fabric, providing a single system, scalability and manageability. According to one embodiment, this virtualization is transparent to the application, and therefore, applications may be scaled to increasing resource demands without modifying the application.

A computer architecture includes a first operating system (COS), which may be a commodity operating system, and a kernel, which acts as a second operating system. The COS is used to boot the system as a whole. After booting, the kernel is loaded and displaces the COS from the system level, meaning that the kernel itself directly accesses predetermined physical resources of the computer. All requests for use of system resources then pass via the kernel. System resources are divided into those that, in order to maximize speed, are controlled exclusively by the kernel, those that the kernel allows the COS to handle exclusively, and those for which control is shared by the kernel and COS. In the preferred embodiment of the invention, at least one virtual machine (VM) runs via a virtual machine monitor, which is installed to run on the kernel. Each VM, the COS, and even each processor in a multiprocessor embodiment, are treated as separately schedulable entities that are scheduled by the kernel. Mechanisms for high-speed I / O between VM's and I / O devices are also included.

Systems and methods for balancing a load among multiple graphics processors that render different portions of a frame. A display area is partitioned into portions for each of two (or more) graphics processors. The graphics processors render their respective portions of a frame and return feedback data indicating completion of the rendering. Based on the feedback data, an imbalance can be detected between respective loads of two of the graphics processors. In the event that an imbalance exists, the display area is re-partitioned to increase a size of the portion assigned to the less heavily loaded processor and to decrease a size of the portion assigned to the more heavily loaded processor.

A system for delivering a fluid to a patient includes a remote controller and an infusion pump. The infusion pump includes a dispenser for dispensing the fluid. An operationally-independent RF telemetry portion is configured to receive an RF data signal from the remote controller. An operationally-independent processing portion is configured to process the RF data signal received by the operationally-independent RF telemetry portion, and control the dispenser in accordance with the RF data signal received by the operationally-independent RF telemetry portion.

A remote direct memory access (RDMA) system is provided in which data is transferred over a network by DMA between from a memory of a first node of a multi-processor system having a plurality of nodes connected by a network and a memory of a second node of the multi-processor system. The system includes a first network adapter at the first node, operable to transmit data stored in the memory of the first node to a second node in a plurality of portions in fulfillment of a DMA request. The first network adapter is operable to transmit each portion together with identifying information and information identifying a location for storing the transmitted portion in the memory of the second node, such that each portion is capable of being received independently by the second node according to the identifying information. Each portion is further capable of being stored in the memory of the second node at the location identified by the location identifying information.

A distributed processing system delegates the allocation and control of computing work units to agent applications running on computing resources including multi-processor and multi-core systems. The distributed processing system includes at least one agent associated with at least one computing resource. The distributed processing system creates work units corresponding with execution phases of applications. Work units can be associated with concurrency data that specifies how applications are executed on multiple processors and / or processor cores. The agent collects information about its associated computing resources and requests work units from the server using this information and the concurrency data. An agent can monitor the performance of executing work units to better select subsequent work units. The distributed processing system may also be implemented within a single computing resource to improve processor core utilization of applications. Additional computing resources can augment the single computing resource and execute pending work units at any time.

A system and method for scheduling jobs in a multiprocessor machine is disclosed. The status of CPUs on node boards in the multiprocessor machine is periodically determined. The status can indicate the number of CPUs available, and the maximum radius of free CPUs available to execute jobs. Memory allocation is also monitored. This information is provided to a scheduler that compares the status of the resources available against the resource requirements of jobs. The node boards and CPUS, as well as other resources such as memory, are arranged in hosts. The scheduler then schedules jobs to hosts that indicate they have resources available to execute the jobs. If none of the hosts indicate they have resources available to execute the jobs, the scheduler will wait until the resources become available. A best fit of job to resources is attained by scheduling jobs to hosts that have the maximum number of free CPUs for a radius corresponding to the CPU radius requirement of a job. Once the job is scheduled to a host, it is dispatched to a host and resources required to execute the job are allocated to the job at the host.

A software application in a multi-processor data router in which a forwarding information base for the router is maintained is provided with a server module and one or more client modules, each client module associated with one or more communication interfaces of the data router. The application is characterized in that the server module sends to each client module only that portion of the forwarding information base specific to the communication interfaces associated with the client module.

A middleware processor provisioning process provisions a plurality of processors in a multi-processor environment. The processing capability of the multiprocessor environment is subdivided and multiple instances of service applications start protected processes to service a plurality of user processing requests, where the number of protected processes may exceed the number of processors. A single processing queue is created for each processor. User processing requests are portioned and dispatched across the plurality of processing queues and are serviced by protected processes from corresponding service applications, thereby efficiently using available processing resources while servicing the user processing requests in a desired manner.

Systems and methods for private addressing in a multi-processor graphics processing subsystem having a number of memories and a number of graphics processors. Each of the memories includes a number of addressable storage locations, and storage locations in different memories may share a common global address. Storage locations are uniquely identifiable by private addresses internal to the graphics processing subsystem. One of the graphics processors is able to access a location in a particular memory by referencing its private address.

In a three-tier ERP implementation, multiple servers are interconnected through one or more network infrastructure. Users may observe poor performance due to the complexity and the number of interconnected components in the implementation. Herein is devised a process for tuning the software component by applying tuning techniques to the OS, SAP application and Database Management System software. For each component, the process identifies potential tuning opportunities of various subcomponents. The process is iterated numerous times through all software components while applying the tuning techniques to derive the most optimal performance for the ERP implementation.

A method for accessing memory by a first processor of a plurality of processors in a multi-processor system includes, responsive to a memory access instruction within a speculative region of a program, accessing contents of a memory location using a transactional memory access to the memory access instruction unless the memory access instruction indicates a non-transactional memory access. The method may include accessing contents of the memory location using a non-transactional memory access by the first processor according to the memory access instruction responsive to the instruction not being in the speculative region of the program. The method may include updating contents of the memory location responsive to the speculative region of the program executing successfully and the memory access instruction not being annotated to be a non-transactional memory access.

A method of operating a multi-processor video encoder by determining a target size corresponding to a preferred number of bits to be used when creating an encoded version of a picture in a group of sequential pictures making up a video sequence. The method includes the steps of calculating a first degree of fullness of a coded picture buffer at a first time, operating on the first degree of fullness to return an estimated second degree of fullness of the coded picture buffer at a second time, and operating on the second degree of fullness to return an initial target sized for the picture. The first time corresponds to the most recent time an accurate degree of fullness of the coded picture buffer can be calculated and the second time occurs after the first time.

A system and a method for failover control comprising: maintaining a primary device table entry (DTE) in a first table activated for a first adapter in communication with a first processor node having a first root complex via a first switch assembly and maintaining a secondary DTE in standby for a second adapter in communication with a second processor node having a second root complex via a second switch assembly; maintaining a primary DTE in a second table activated for the second adapter and maintaining a secondary DTE in standby for the first adapter; and upon a failover, updating the secondary DTE in the first table as an active entry for the second adapter and forming a path to enable traffic to route from the second adapter through the second switch assembly over to the first switch assembly and up to the first root complex of the first processor node.

An enhanced memory algorithmic processor ("MAP") architecture for multiprocessor computer systems comprises an assembly that may comprise, for example, field programmable gate arrays ("FPGAs") functioning as the memory algorithmic processors. The MAP elements may further include an operand storage, intelligent address generation, on board function libraries, result storage and multiple input / output ("I / O") ports. The MAP elements are intended to augment, not necessarily replace, the high performance microprocessors in the system and, in a particular embodiment of the present invention, they may be connected through the memory subsystem of the computer system resulting in it being very tightly coupled to the system as well as being globally accessible from any processor in a multiprocessor computer system.

A virtual routing platform includes a line interface a plurality of virtual routing engines (VREs) to identify packets of different packet flows and perform a hierarchy of metering including at least first and second levels of metering on the packet flows. A first level of metering may be performed on packets of a first packet flow using a first metering control block (MCB). The first level of metering may be one level of metering in a hierarchy of metering levels. A second level of metering on the packets of the first packet flow and packets of a second flow using a second MCB. The second level of metering may be another level of metering in the hierarchy. A cache-lock may be placed on the appropriate MCB prior to performing the level of metering. The first and second MCBs may be data structures stored in a shared memory of the virtual routing platform. The cache-lock may be released after performing the level of metering using the MCB. The cache-lock may comprise setting a lock-bit of a cache line index in a cache tag store, which may identify a MCB in the cache memory. The virtual routing platform may be a multiprocessor system utilizing a shared memory having a first and second processors to perform levels of metering in parallel. In one embodiment, a virtual routing engine may be shared by a plurality of virtual router contexts running in a memory system of a CPU of the virtual routing engine. In this embodiment, the first packet flow may be associated with one virtual router context and the second packet flow is associated with a second virtual router context. The first and second routing contexts may be of a plurality of virtual router contexts resident in the virtual routing engine.

A cache memory shared among a plurality of separate, disjoint entities each having a disjoint address space, includes a cache segregator for dynamically segregating a storage space allocated to each entity of the entities such that no interference occurs with respective ones of the entities. A multiprocessor system including the cache memory, a method and a signal bearing medium for storing a program embodying the method also are provided.

A computer system including a first processor, a second processor in communication with the first processor, a memory coupled to the first and second processors (i.e., a shared memory) and including multiple memory locations, and a storage device coupled to the first processor. The first and second processors implement virtual memory using the memory. The first processor maintains a first set of page tables and a second set of page tables in the memory. The first processor uses the first set of page tables to access the memory locations within the memory. The second processor uses the second set of page tables, maintained by the first processor, to access the memory locations within the memory. A virtual memory page replacement method is described for use in the computer system, wherein the virtual memory page replacement method is designed to help maintain paged memory coherence within the multiprocessor computer system.

A Cellular Multi-Processor Serve provides partitions having different Operating Systems such as Windows, Unix OS2200 (Unisys), Master Control Program (Unisys) or other Operating Systems, which could be designated as OS-A or OS-B. The present method and system collects and scans availability and reliability information which involve non-windows partitions with respect to planned and unplanned stops, system starts and different categories of error conditions.

A processor architecture containing multiple closely coupled processors in a form of symmetric multiprocessing system is provided. The special coupling mechanism allows it to speculatively execute multiple threads in parallel very efficiently. Generally, the operating system is responsible for scheduling various threads of execution among the available processors in a multiprocessor system. One problem with parallel multithreading is that the overhead involved in scheduling the threads for execution by the operating system is such that shorter segments of code cannot efficiently take advantage of parallel multithreading. Consequently, potential performance gains from parallel multithreading are not attainable. Additional circuitry is included in a form of symmetrical multiprocessing system which enables the scheduling and speculative execution of multiple threads on multiple processors without the involvement and inherent overhead of the operating system. Advantageously, parallel multithreaded execution is more efficient and performance may be improved.

A method of maintaining cache coherency, by designating one cache that owns a line as a highest point of coherency (HPC) for a particular memory block, and sending a snoop response from the cache indicating that it is currently the HPC for the memory block and can service a request. The designation may be performed in response to a particular coherency state assigned to the cache line, or based on the setting of a coherency token bit for the cache line. The processing units may be grouped into clusters, while the memory is distributed using memory arrays associated with respective clusters. One memory array is designated as the lowest point of coherency (LPC) for the memory block (i.e., a fixed assignment) while the cache designated as the HPC is dynamic (i.e., changes as different caches gain ownership of the line). An acknowledgement snoop response is sent from the LPC memory array, and a combined response is returned to the requesting device which gives priority to the HPC snoop response over the LPC snoop response.

An atomic transaction includes one or more memory access operations that are completed atomically. A Best-Effort Transaction (BET) system makes its best effort to complete each atomic transaction without guaranteeing completion of all atomic transactions. When an atomic transaction is aborted, BET may provide software with appropriate runtime information such as cause of the abortion. With proper coherence layer enhancements, BET can be implemented efficiently for multiprocessor systems, using caches as buffers for data accessed by atomic transactions. Furthermore, with appropriate fairness support, forward progress can be guaranteed for atomic transactions that incur no buffer overflow.

An enhanced SCSI storage adapter with multiple queues for use by different server processors or partitions. For a non-partitioned server, the operating system (OS) owns the SCSI storage adapter, controls the adapter queues, both creation of and changes to the queues, and updates the queue table(s) in the storage adapter with queue address information, device list, message signaled interrupt (MSI) information and optional queue priorities. An OS operator can specify that one or more SCSI devices can be accessed by a specific processor or group of processors. The processor or group of processors is given an adapter queue to access the SCSI device or devices. For a partitioned server, one partition, which may be a hosting partition, owns the SCSI storage adapter, controls the adapter queues, both creation of and changes to the queues, and updates the queue table(s) in the storage adapter with queue address information, device list, message signaled interrupt (MSI) information and optional queue priorities. A system operator can assign one or more SCSI devices under a storage adapter to a partition. Each partition that has access to a SCSI device(s) under a SCSI adapter is given an adapter queue to access the device(s).

Improved method and apparatus for parallel processing. One embodiment provides a multiprocessor computer system that includes a first and second node controller, a number of processors being connected to each node controller, a memory connected to each controller, a first input / output system connected to the first node controller, and a communications network connected between the node controllers. The first node controller includes: a crossbar unit to which are connected a memory port, an input / output port, a network port, and a plurality of independent processor ports. A first and a second processor port connected between the crossbar unit and a first subset and a second subset, respectively, of the processors. In some embodiments of the system, the first node controller is fabricated onto a single integrated-circuit chip. Optionally, the memory is packaged on plugable memory / directory cards wherein each card includes a plurality of memory chips including a first subset dedicated to holding memory data and a second subset dedicated to holding directory data. Further, the memory port includes a memory data port including a memory data bus and a memory address bus coupled to the first subset of memory chips, and a directory data port including a directory data bus and a directory address bus coupled to the second subset of memory chips. In some such embodiments, the ratio of (memory data space) to (directory data space) on each card is set to a value that is based on a size of the multiprocessor computer system.