摘要:
A method of detecting and recovering from data corruption of a database is characterized by the step of protecting data of the database with codewords, one codeword for each region of the database; and verifying that a codeword matches associated data before the data is read from the database to prevent transaction-carried corruption. A deferred maintenance scheme is recommended for the codewords protecting the database such that the method of detecting and recovering from data corruption of a database may comprise the steps of protecting data of the database with codewords, one codeword for each region of the database; and asynchronously maintaining the codewords to improve concurrency of the database. Moreover, the database may be audited by using the codewords and noting them in a table and protecting regions of the database with latches. Once codeword values are computed and checked against noted values in memory, a flush can cause codewords from outstanding log records to be applied to the stored codeword table.
摘要:
For use with a database of data records organized into components, the database stored in a memory, a processing system for, and method of, physically versioning the database. In one embodiment, the processing system includes: (1) a component copier that creates a physical copy of an original component to be affected by an update transaction to be applied to the database, and that causes pointers in nodes of the physical copy to point to other nodes in the physical copy, (2) a data updater, associated with the component copier, that applies the update transaction to the physical copy to create therefrom a new physical version, the original component remaining unaffected by the update transaction and (3) a pointer updater, associated with the data updated, that employs an atomic word write to revise a component pointer, associated with the database, to cause the pointer to point to the new physical version.
摘要:
A continuous media server that provides support for the storage and retrieval of continuous media data at guaranteed rates using one of two fault-tolerant approaches that rely on admission control in order to meet rate guarantees in the event of a failure of the data storage medium that renders part of the continuous media inaccessible. In the first approach, a declustered parity storage scheme is used to uniformly distribute the additional load caused by a disk failure, uniformly across the disks. Contingency bandwidth for a certain number of clips is reserved on each disk in order to retrieve the additional blocks. In the second approach, data blocks in a parity group are prefetched and thus in the event of a disk failure only one additional parity block is retrieved for every data block to be reconstructed. While the second approach generates less additional load in the event of a failure, it has higher buffer requirements. For the second approach, parity blocks can either be stored on a separate parity disk, or distributed among the disks with contingency bandwidth reserved on each disk.
摘要:
A method and an apparatus are disclosed for providing enhanced pay per view in a video server. Specifically, the present invention periodically schedules a group of non pre-emptible tasks corresponding to videos in a video server having a predetermined number of processors, wherein each task is defined by a computation time and a period. To schedule the group of tasks, the present invention divides the tasks into two groups according to whether they may be scheduled on less than one processor. The present invention schedules each group separately. For the group of tasks scheduleable on less than one processor, the present invention conducts a first determination of scheduleability. If the first determination of scheduleability deems the group of tasks not scheduleable, then the present invention conducts a second determination of scheduleability. If the second determination of scheduleability also deems the group of tasks not scheduleable, then the present invention recursively partitions the group of tasks in subsets and re-performs the second determination of scheduleability. Recursive partitioning continues until the group of tasks is deemed scheduleable or no longer partitionable. In the latter case, the group of tasks is deemed not scheduleable.
摘要:
Two methods are disclosed for storing multimedia data that reduces the amount of disk I/O required by the system and cache misses experienced by the system. The first method determines the future access of each data buffer in a cache memory. Once the future of the data buffer is determined, the data buffer with the maximum future is allocated to store new blocks of data. The method approximates an optimal method of data buffer allocation, by calculating the future of a data buffer, relative to clients that will access the data buffers. The second method orders the clients based on the increasing distance of each client from the previous client; clients release the buffers in this order into a LIFO queue; if a buffer is selected to load a new block of data, the buffer at the head of the LIFO queue is selected.
摘要:
Buffer space and disk bandwidth resources in a continuous media server are continuously re-allocated in order to optimize the number of continuous media requests which may be concurrently serviced at guaranteed transfer rates using on demand paging. Disk scheduling is provided to ensure that whenever an admitted request references a page of data, the page is available in a buffer for transfer to a client. Data for continuous media data files are stored on disk or held in the buffer to eliminate disk bandwidth limitations associated with concurrently servicing any number or combination of requests, provided buffer space is sufficient. Multiple requests for continuous media data files are selectively included in groups for servicing in order to provide that buffer and disk bandwidth requirements are maintained at a minimum and within available resource capabilities.
摘要:
Disclosed is a mobile station having an expanded storage space, and a method of retrieving a file stored in a remote storage server. The method may include determining whether the file is locally stored or remotely stored, deleting files in a local storage to provide space if the file is determined to reside in a remote storage server, downloading the file from a remote storage server over the Internet by prefetching a portion of the file from the remote storage server, and opening the prefetched portion of the file while concurrently downloading a remaining portion of the file from the remote storage server.
摘要:
Restoration is provided in a communication system having two or more VPN endpoints coupled together through a network comprising a number of nodes interconnected through edges. VPN endpoints coupled through the network are adapted to communicate through a single connection with multiple other VPN endpoints. The single connection may be a VPN hose connection. A restoration topology, comprising backup edges corresponding to primary edges in the network, is determined for the network. A given primary edge is replaced with one or more backup edges if the given primary edge fails. A graph may represent the network and a tree may represent the connections in the network for VPNs connecting the VPN endpoints. The graph can be reduced to a second graph by determining shortest paths between each node in the tree and creating the backup edges from the shortest paths. The second graph can be reduced to a third graph by adding additional backup edges from tree nodes having non-tree edges to least common ancestor nodes. The third graph can be used to create a two-edge connected tree.
摘要:
The invention comprises a method and apparatus for determining a rank of a query value. Specifically, the method comprises receiving a rank query request, determining, for each of the at least one remote monitor, a predicted lower-bound rank value and upper-bound rank value, wherein the predicted lower-bound rank value and upper-bound rank value are determined according to at least one respective prediction model used by each of the at least one remote monitor to compute the at least one local quantile summary, computing a predicted average rank value for each of the at least one remote monitor using the at least one predicted lower-bound rank value and the at least one predicted upper-bound rank value associated with the respective at least one remote monitor, and computing the rank of the query value using the at least one predicted average rank value associated with the respective at least one remote monitor.
摘要:
The invention improves query response latency in a peer-to-peer network. The invention augments core neighbors included in a neighbor list of a peer node by selecting auxiliary neighbors for inclusion in the neighbor list of the peer node of a peer-to-peer network. In one embodiment, a method includes maintaining query frequency information associated with each of a plurality of peer nodes of the peer-to-peer network, selecting at least one of the peer nodes of the peer-to-peer network as an auxiliary neighbor using the query frequency information, and updating a neighbor list to include the at least one peer node selected as the at least one auxiliary neighbor. The core neighbors stored at a peer node attempt to minimize the worst-case query latency for that peer node. The auxiliary neighbors stored at the peer node attempt to reduce the average query latency for that peer node. The auxiliary neighbors may be used in any peer-to-peer system, such as Pastry, Tapestry, Chord, SkipGraph, and the like.