4168results about "Program synchronisation" patented technology

A method and system for leasing storage locations in a distributed processing system is provided. Consistent with this method and system, a client requests access to storage locations for a period of time (lease period) from a server, such as the file system manager. Responsive to this request, the server invokes a lease period algorithm, which considers various factors to determine a lease period during which time the client may access the storage locations. After a lease is granted, the server sends an object to the client that advises the client of the lease period and that provides the client with behavior to modify the lease, like canceling the lease or renewing the lease. The server supports concurrent leases, exact leases, and leases for various types of access. After all leases to a storage location expire, the server reclaims the storage location.

A novel massively parallel supercomputer of hundreds of teraOPS-scale includes node architectures based upon System-On-a-Chip technology, i.e., each processing node comprises a single Application Specific Integrated Circuit (ASIC). Within each ASIC node is a plurality of processing elements each of which consists of a central processing unit (CPU) and plurality of floating point processors to enable optimal balance of computational performance, packaging density, low cost, and power and cooling requirements. The plurality of processors within a single node may be used individually or simultaneously to work on any combination of computation or communication as required by the particular algorithm being solved or executed at any point in time. The system-on-a-chip ASIC nodes are interconnected by multiple independent networks that optimally maximizes packet communications throughput and minimizes latency. In the preferred embodiment, the multiple networks include three high-speed networks for parallel algorithm message passing including a Torus, Global Tree, and a Global Asynchronous network that provides global barrier and notification functions. These multiple independent networks may be collaboratively or independently utilized according to the needs or phases of an algorithm for optimizing algorithm processing performance. For particular classes of parallel algorithms, or parts of parallel calculations, this architecture exhibits exceptional computational performance, and may be enabled to perform calculations for new classes of parallel algorithms. Additional networks are provided for external connectivity and used for Input / Output, System Management and Configuration, and Debug and Monitoring functions. Special node packaging techniques implementing midplane and other hardware devices facilitates partitioning of the supercomputer in multiple networks for optimizing supercomputing resources.

Methods for load rebalancing by clients in a network are disclosed. Client load rebalancing allows the clients to optimize throughput between themselves and the resources accessed by the nodes. A network, which implements this embodiment of the invention, can dynamically rebalance itself to optimize throughput by migrating client I / O requests from over_utilized pathways to under_utilized pathways. Client load rebalancing allows a client to re-map a path through a plurality of nodes to a resource. The re-mapping may take place in response to a redirection command from an overloaded node.

Database data is maintained reliably and invalidated based on actual changes to data in the database. Updates or changes to data are detected without parsing queries submitted to the database. The dependencies of a query can be determined by submitting a version of the received query to the database through a native facility provided by the database to analyze how query structures are processed. The caching system can access the results of the facility to determine the tables, rows, or other partitions of data a received query is dependent upon or modifies. An abstracted form of the query can be cached with an indication of the tables, rows, etc. that queries of that structure access or modify. The tables a write or update query modifies can be cached with a time of last modification. When a query is received for which the results are cached, the system can readily determine dependency information for the query, the last time the dependencies were modified, and compare this time with the time indicated for when the cached results were retrieved. By passing versions of write queries to the database, updates to the database can be detected.

Methods, apparatus, and products are disclosed for processing data access requests among a plurality of compute nodes. One compute node operates as a processing node, and one compute nodes operates as a requesting node. The processing node receives, from the requesting node, a data access request to access data currently being processed by the processing node. The processing node also receives, from the requesting node, a processing directive. The processing directive specifies data processing operations to be performed on the data specified by the data access request. The processing node performs, on behalf of the requesting node, the data processing operations specified by the processing directive on the data specified by the data access request. The processing node transmits, to the requesting node, results of the data processing operations performed on the data by the processing node on behalf of the requesting node.

A method, apparatus and computer program product are provided to synchronize multiple devices. In regards to a method, an indication is received that a view of a task is presented by a first device. The method causes state information to be provided to a second device to permit the second device to be synchronized with the first device and to present a view of the task, either the same view or a different view than that presented by the first device. The method also receives information relating to a change in state of the task that is provided by one of the devices while a first view of the task is presented thereupon. Further, the method causes updated state information to be provided to another one of the devices to cause the other device to remain synchronized and to update a second view of the task that is presented.

An apparatus, system, and method are disclosed for data storage with progressive redundant array of independent drives (“RAID”). A storage request receiver module, a striping module, a parity-mirror module, and a parity progression module are included. The storage request receiver module receives a request to store data of a file or of an object. The striping module calculates a stripe pattern for the data. The stripe pattern includes one or more stripes, and each stripe includes a set of N data segments. The striping module writes the N data segments to N storage devices. Each data segment is written to a separate storage device within a set of storage devices assigned to the stripe. The parity-mirror module writes a set of N data segments to one or more parity-mirror storage devices within the set of storage devices. The parity progression module calculates a parity data segment on each parity-mirror device in response to a storage consolidation operation, and stores the parity data segments. The storage consolidation operation is conducted to recover storage space and / or data on a parity-mirror storage device.

A hardware-assisted Distributed Memory System may include software configurable shared memory regions in the local memory of each of multiple processor cores. Accesses to these shared memory regions may be made through a network of on-chip atomic transaction engine (ATE) instances, one per core, over a private interconnect matrix that connects them together. For example, each ATE instance may issue Remote Procedure Calls (RPCs), with or without responses, to an ATE instance associated with a remote processor core in order to perform operations that target memory locations controlled by the remote processor core. Each ATE instance may process RPCs (atomically) that are received from other ATE instances or that are generated locally. For some operation types, an ATE instance may execute the operations identified in the RPCs itself using dedicated hardware. For other operation types, the ATE instance may interrupt its local processor core to perform the operations.

A novel linked-list-based concurrent shared object implementation has been developed that provides non-blocking and linearizable access to the concurrent shared object. In an application of the underlying techniques to a deque, non-blocking completion of access operations is achieved without restricting concurrency in accessing the deque's two ends. In various realizations in accordance with the present invention, the set of values that may be pushed onto a shared object is not constrained by use of distinguishing values. In addition, an explicit reclamation embodiment facilitates use in environments or applications where automatic reclamation of storage is unavailable or impractical.

A technique for utilizing resources in a virtual machine operating system. The virtual machine operating system comprises a multiplicity of virtual machines. A share of resources is allocated to each of the virtual machines. Utilization by one of the virtual machines of the resources allocated to the one virtual machine is automatically monitored. If the one virtual machine needs additional resources, the one virtual machine is automatically cloned. The clone is allocated a share of the resources taken from the shares of other of the virtual machines, such that the resultant shares allocated to the one virtual machine and the clone together are greater than the share allocated to the one virtual machine before the one virtual machine was cloned. The clone performs work with its resources that would have been performed by the one virtual machine if not for the existence of said clone.

A method and apparatus for dynamic lock granularity escalation and de-escalation in a computer system is provided. Upon receiving a request for a resource, a scope of a previously granted lock is modified. According to one embodiment, hash lock de-escalation is employed. In hash lock de-escalation, the scope of the previously granted lock held on a set of resources is reduced by de-escalating the previously granted lock from a coarser-grain lock to one or more finer-grain locks on members of the set. According to another embodiment, hash lock escalation is employed. In hash lock escalation, the scope of previously granted locks held on one or more members of the set of resources are released and promoted into a coarser-grain lock that covers the set of resources as well as the requested resource.