119results about How to "Avoid overhead" patented technology

The present invention discloses an efficient architecture for routing in a very large autonomous system where many of the layer 3 routers are attached to a common connection-oriented layer 2 subnetwork, such as an ATM network. In a preferred embodiment of the invention, a permanent topology of routers coupled to the subnetwork is connected by permanent virtual circuits. The routers can further take advantage of both intra-area and inter-area shortcuts through the layer 2 network to improve network performance. The routers pre-calculate shortcuts using information from link state packets broadcast by other routers and store the shortcuts to a given destination in a forwarding table, along with corresponding entries for a next hop along the permanent topology. The present invention allows the network to continue to operate correctly if layer 2 resource limitations preclude the setup of additional shortcuts.

The present invention describes a method, system and an interconnecting handler for sending connection-oriented or connectionless data between two endpoints in a protocol architecture. The transport protocol in the protocol architecture is in a preferred embodiment the Stream Control Transmission Protocol (SCTP). The present invention allows the setup and release of a connection when using a simplified protocol stack which is able to provide the same kind of services without using peer-to-peer messages like SCCP or SUA use. The present invention also enables discrimination between connection-oriented and connectionless services without peer-to-peer signalling.

In a virtual machine environment, a method and apparatus for the use of multiple heaps to retain persistent data and transient data wherein the multiple heaps enables a single virtual machine to be easily resettable, thus avoiding the need to terminate and start a new Virtual Machine as well as enabling a single virtual machine to retain data and objects across multiple applications, thus avoiding the computing resource overhead of relinking, reloading, reverifying, and recompiling classes. The memory hierarchy includes a System Heap, a Middleware Heap and a Transient Heap. The use of three heaps enables garbage collection to be selectively targeted to one heap at a time in between applications, thus avoiding this overhead during the life of an application.

A traditional structured data store is leveraged to provide the benefits of an unstructured full-text search system. A fixed number of “extended” columns is added to the traditional structured data store to form an “enhanced structured data store” (ESDS). The extended columns are independent of any regular columnar interpretation of the data and enable the data that they store to be searched using standard full-text query syntax/techniques that can be executed faster (as opposed to SQL syntax). In other words, the added columns act as a search index. A token is stored in an appropriate extended column based on that token's hash value. The hash value is determined using a hashing scheme, which operates based on the value of the token, rather than the meaning of the token. This enables subsequent searches to be expressed as full-text queries without degrading the ensuing search to a brute force scan.

A query facility for database queries dynamically determines whether selective portions of a database table are likely to benefit from separate query execution strategies, and constructs an appropriate separate execution strategies accordingly. Preferably, the database contains at least one relatively large table comprising multiple partitions, each sharing the definitional structure of the table and containing a different respective discrete subset of the table records. The query facility compares metadata for different partitions to determine whether sufficiently large differences exist among the partitions, and in appropriate cases selects one or more partitions for separate execution strategies. Preferably, partitions are ranked for separate evaluation using a weighting formula which takes into account: (a) the number of indexes for the partition, (b) recency of change activity, and (c) the size of the partition.

Within a SIMD processor 2 data processing instructions are provided which specify parallel lanes of processing to be performed upon respective data elements. The data elements are permitted to vary in size whilst the number of processing lanes remain constant. Thus, the destination register size for a multiplication may be double the source register size.

For a host that executes one or more guest virtual machines (GVMs), some embodiments provide a novel encryption method for encrypting the data messages sent by the GVMs. The method initially receives a data message to send for a GVM executing on the host. The method then determines whether it should encrypt the data message based on a set of one or more encryption rules. When the process determines that it should encrypt the received data message, it encrypts the data message and forwards the encrypted data message to its destination; otherwise, the method just forwards the received data message unencrypted to its destination. In some embodiments, the host encrypts differently the data messages for different GVMs that execute on the host. When two different GVMs are part of two different logical overlay networks that are implemented on common network fabric, the method in some embodiments encrypts the data messages exchanged between the GVMs of one logical network differently than the data messages exchanged between the GVMs of another logical network. In some embodiments, the method can also encrypt different types of data messages from the same GVM differently. Also, in some embodiments, the method can dynamically enforce encryption rules in response to dynamically detected events, such as malware infections.

A computer architecture for avoiding a deadlock condition in an interconnection network comprises a messaging buffer having a size pre-calculated to temporarily store outgoing messages from a node. Messages are classified according to their service requirements and messaging protocols, and reserved quotas in the messaging buffer are allocated for different types of messages. The allocations of the reserved quotas are controlled by a mechanism that, to prevent overflow, limits the maximum number of messages that can be outstanding at any time. The messaging buffer is sized large enough to guarantee that a node is always able to service incoming messages, thereby avoiding deadlock and facilitating forward progress in communications. The buffer may be bypassed to improve system performance when the buffer is empty or when data in the buffer is corrupted. In addition, a multicast engine facilitates dense packing of the buffer and derives information from a message header to determine whether there is a multicast to perform and to permit passage of messages. Other considerations to reduce the buffer size are incorporated.

A system and method for managing network devices using a metadata gateway. The metadata gateway provides translation of metadata to and from a database format and Interface Definition Language (IDL), which is operable across a plurality of platforms and across a plurality of programming languages. Metadata may be retrieved through the metadata gateway by a client manager application sending a request for type information for a managed object attribute or event in IDL through a CORBA Object Request Broker (ORB) to the metadata gateway, which then reads the type information from a metadata repository, where the type information is stored in a database format. The metadata gateway then translates the retrieved type information from the database format to IDL and sends the translated type information to the ORB, which sends the translated type information for the attribute or event to the client manager application. Metadata may be encoded through the metadata gateway by sending the metadata in IDL to the metadata gateway, which translates the type information from IDL to a database format and stores the type information in the metadata repository.

A method of allocating a radio resource is provided. The method includes: receiving space division multiple access (SDMA) information for downlink transmission; transmitting a result of channel estimation performed on channels corresponding to data streams transmitted in downlink according to the SDMA information; and receiving the data streams through the respective channels according to the result of channel estimation. Accordingly, radio resource request states of stations can be collectively considered.

A method and system is provided for splitting a print job into its preamble and at least one chunk. The splitter maintains a collection of RIP node addresses to which chunks of the job currently being split have already been sent. When a new chunk is about to be sent, the splitter checks whether each RIP node address has already received a chunk. If the RIP node has not already received a chunk, the splitter sends the preamble as well as the chunk to an available RIP associated with the RIP node. If, however, the RIP node address has already received a chunk, only the portion of the chunk after the preamble is sent to an available RIP associated with the RIP node and communicate the location of the preamble to the available RIP node. The preamble may contain common content for each job.

The invention discloses a large-scale data field volume rendering method based on partition strategy, mainly solving the problem of small storage capacity of the volume rendering method based on GPU. The invention has the follow rendering process: whole volume data is decomposed to generate a plurality of subblocks with the same size, and all generated subblocks are numbered; distance between each subblock and a sight plane is calculated, and the subblocks are arranged from back to front according to the distance; octree is used for organizing the subblocks, and is used for removing the volume data in invalid subblocks; the subblocks are successively loaded into a texture memory in order; the subblocks are successively rendered by a 3D texture volume rendering algorithm based on GPU so as to obtain the rendering result of each subblock; the rendering results of all subblocks are mixed from back to front in a frame buffer to obtain a whole rendering image. The invention has the advantages of small cost for loading the volume data into the texture memory, and high speed of treating and rendering volume data, and can be used for quickly rendering texture memory capacity volume data whose data scale is bigger than GPU.

The invention provides a splitting method for a search string of Chinese vertical search, which is to split the search string of the Chinese vertical search through an entity dictionary and an unsupervised learning method. The splitting method for the search string of Chinese vertical search comprises the steps of: establishing an entity dictionary and a language model; carrying out entity name matching for the search string; processing non-Chinese characters in the search string; performing word segmentation on the search string; establishing a weight matrix of a candidate phrase; obtaining a weight of a combination of all candidate phrases of the search string; regarding a phrase combination with the maximum weight as a splitting result of the search string to be turned back. The method overcomes the difficulty of a dictionary method for dealing with ambiguity meanings, avoids expense of a supervised learning method taken for manual corpus tagging and reduces influence from noise to segmentation boundary without the supervised learning method.

The invention provides a memory computing-based customizable multimode big data processing system. The system comprises a data storage layer module, a memory-based data sharing and management layer module, a memory computing-based universal execution framework layer module and an access interface layer module, wherein a distributed memory abstraction mechanism, a position-sensing scheduling mechanism and a distributed hybrid column storage mechanism of data are adopted. According to the system, massive data storage is constructed; a cluster concurrent computing-oriented memory data management and sharing framework is provided, and an efficient and customized big data multimode universal processing framework is provided; batch processing and real-time data stream computing are supported; and the support is provided for flexible analysis and deep utilization of the data.

A method for configuring radio access functionality of a wireless communication unit within a wireless communication network, the method comprising sending to the wireless communication unit an indication of a required radio access protocol version to which the wireless communication unit is to be reconfigured.

The invention discloses a KCF-based improved anti-shielding target tracking method. The KCF-based improved anti-shielding target tracking method is based on a KCF algorithm and combined with a Kalmanfilter. The KCF algorithm has the advantages of fast tracking speed, high tracking precision and wide application in the target tracking field, but cannot cope with situations of scale changes and target large-area shielding. The KCF-based improved anti-shielding target tracking method adopts parallel computing to realize multi-scale target detection of a KCF, and designs an effective algorithm for determining target shielding. When determining that the target is severely shielded, the KCF-based improved anti-shielding target tracking method adopts a prediction result of the Kalman filter as acurrent frame target position. Experiments prove that the KCF-based improved anti-shielding target tracking method can effectively cope with the occurrence of tracking drift situation when the KCF algorithm encounters target large-area shielding, so that the KCF algorithm can track the target more precisely when encountering the shielding.

The invention provides a query method based on regional bitmap indexes in a cloud environment. The method comprises the following steps of: 1) establishing the regional bitmap indexes; 1.1) performing range division on index attributes on a data table in the cloud environment to generate a global sequencing table of attribute values, wherein the global sequencing table is used for sequencing tuples by using a set rule; 1.2) establishing an indicating bitmap on each data node according to the range division result, wherein the indicating bitmap records the storage condition of local attribute values; 1.3) establishing a local bitmap index on each data node according to the framework of the cloud environment to finish establishment of the regional bitmap indexes; and 2) inputting a query condition, establishing a condition bitmap according to the query condition by a main node, and distributing the condition bitmap to each data node, wherein the condition bitmap covers all probabilities included in the query condition; and concurrently executing retrieval task through each data node, acquiring the query result of each data node by the main node, and returning a union set of the query results of the data nodes to a user. By establishing the regional bitmap indexes, configurable parallel computing resources in the cloud environment can be fully utilized, and quick response can be provided for the data query request using capacity comparison as a condition.

The invention discloses a method and a system for estimating the capacity of a server. The method disclosed by the invention comprises the following steps of: counting the business-processing capacity finished by the server within unit time and the time length at the peak period each day; calculating the percent of the processing capacity of the server at the peak period accounting for the all-day processing capacity and the complexity parameter; adjusting the redundancy factor of the server and the utilization ratio of a central processor; and estimating the capacity of the server according to the business-processing capacity finished by the server within the unit time, the time length at the peak period each day, the percent of the processing capacity at the peak period accounting for the all-day processing capacity, the complexity parameter, the redundancy factor and the utilization ratio of the central processor. The invention also discloses the system for estimating the capacity of the server. Through the introduction of the complexity parameter in the invention, the differences of different application systems are eliminated, and a universal estimating method for the capacity of the server of each application system is realized, so that the estimating efficiency is improved; and estimation is carried out by utilizing the actual data provided by the current online application system, so that errors between the estimated capacity and the actual capacity are reduced.

The invention discloses a high-performance mass storage system and a high-performance mass storage method, and belongs to the technical field of storage. The mass storage system comprises a file server and a network disk array group, wherein the network disk array group comprises a plurality of network disk arrays, all network disk arrays are mounted onto the file server through a peripheral channel in a DAS mode, and then, all network disk arrays and the file server are connected onto a switching network through network channels; a network user is connected onto the switching network, and realizes interaction with the file server and the network disk arrays through the switching network; the file server is in charge of monitoring all operations in the mass storage system and provides data description for the network user; the network disk arrays are in charge of the data access of the network user; and the peripheral channel is an SCSI bus or an FC bus or a network passage. The high-performance mass storage system and the high-performance mass storage method have the beneficial effects that the advantages of a centralized storage system and a distributed storage system are simultaneously achieved, and the file centralized management and the data distributed storage are realized.

The invention discloses methods and devices for storing and searching data. The method for storing data comprises the steps of: classifying and storing performance data acquired in every data acquisition period according to service characteristics in different performance data classes in a performance file; and storing the index information corresponding to every performance data class in a class index file corresponding to the performance file. The method for searching the data comprises the steps of: determining the performance data class to which to-be-searched performance data belongs according to the to-be-searched performance data; searching the index information corresponding to the performance data class in the class index file; and searching the to-be-searched performance data in the performance file according to the index information. According to the method and the device disclosed by the invention, the performance data can be rapidly searched with a high inquiry efficiency, IO (input/output) operation is reduced, the search complexity is decreased, and the performance of disk random reading and writing is improved, thereby the system availability is improved.

Techniques for handling exceptions raised during the execution of Java computer programs are disclosed. The techniques can be used by a Java virtual machine to efficiently handle exceptions. A method descriptor is implemented in a Java method frame which is stored in the Java execution stack. The method descriptor provides one or more references to exception handlers associated with the Java method. The references can be used to quickly identify and invoke the appropriate exception handler. This can be achieved without having to use a native language execution stack. As a result, the overhead associated with several returns from native functions (routines or methods) can be avoided since the information needed to invoke the appropriate exception handler can be obtained efficiently from the Java execution stack. Accordingly, the performance of Java virtual machines, especially those operating with limited resources, can be significantly enhanced.

The invention discloses a data transmission method and a data transmission system, which are used for solving the technical problems of uplink transmission conflicts or uplink resource wasting under the condition of not introducing a relay-physical hybrid automatic repeat request indicator channel (R-PHICH) in a wireless relay network system. In the method and the system, a first network element explicitly and/or implicitly indicates hybrid automatic repeat request (HARQ) feedback information to a second network element; and the second network element transmits data or does not transmit the data according to the HARQ feedback information. By the method and the system, the data transmission conflicts are avoided, and resource scheduling needs are satisfied by an explicit and/or implicit indication way; and the R-PHICH is not required so as to avoid overhead in a Un interface resource configuration PHICH, reduce the signaling overhead of a relay-physical downlink control channel (R-PDCCH) for a Un interface, improve scheduling efficiency and increase the utilization rate of the resources.

The invention relates to the technical field of Internet massive data storage and discloses a user data storage method. The method includes: storing user data in the file body of a pre-created basic file to allow the user data of the same user to be stored continuously, wherein the basic file stores the user data of preset number of users; recording the location indexes of the storage locations of the user data in the basic file in the file header of the basic file. The invention further discloses a user data storage system, and a user data reading method and system. The method has the advantages the user data of preset number of users is stored in the basic file continuously, and data inquiry or reading efficiency is increased; first data node information maintenance of all users in the prior art can be avoided, extra internal memory resource expense is avoided, and maintenance and operation costs are lowered.

The present invention discloses a multi-CPU communication method and a system thereof. The method is achieved through PCI Express bus equipment, which includes: a. a source CPU and a target CPU respectively carry out the initial configuration of the NT bridge of PCI Express bus equipment on the side and then step b is carried out after the initial configuration on any side is completed; b. the CPU on the side where the initial configuration is completed starts the periodic test for the connection of PCI Express bus between the source CPU and the target CPU; the source CPU needs to send data messages to the target CPU; if the connection is normal, the periodic test ends to enter step c and if the connection is interrupted, the periodic test continues; c. the source CPU sends data messages to the target CPU according to the initial configuration of the NT bridge; d. the target CPU receives the data messages. The technical proposal of the present invention achieves the high-speed communication among multiple CPUs.

The invention relates to a heterogeneous multi-core thread scheduling method. The heterogeneous multi-core thread scheduling method includes: respectively generating a ranking list for threads and cores according to dynamic characteristics of a program, finding out optimal stable match of the threads and the cores according to the ranking lists, and performing thread scheduling according to stable match. The heterogeneous multi-core thread scheduling method specifically includes: receiving characteristic vectors of the threads running in the cores, and selecting priority ranking of the cores for the threads according to the characteristic vectors; ranking each thread for each core; receiving the ranking lists of the threads and the cores, and finding out stable match results of the threads and the cores; and receiving the stable match results, scheduling by an operating system and allocating each thread to the corresponding core for running. Huge expenditure caused by sampling scheduling is avoided, the heterogeneous multi-core thread scheduling method takes more complex factors influencing performances and power consumption into consideration, and only relative relations rather than specific values need to be predicated, so that model complexity is lowered while scheduling precision is also improved.

The invention discloses a bidirectional forwarding detection system and a detection time configuration method of bidirectional forwarding detection. In the process of realizing bidirectional forwarding detection, the network time delay between a bidirectional forwarding detection local end and a bidirectional forwarding detection opposite end is tested, and then the detection time of the bidirectional forwarding detection local end and the detection time of the bidirectional forwarding detection opposite end are configured to be the same based on the measured network time delay; that is, according to the invention, the detection time of bidirectional forwarding detection is configured based on the network time delay between the bidirectional forwarding detection local end and the bidirectional forwarding detection opposite end and is not configured only according to the packet transmitting and receiving interval and the detection times configured on the network equipment on two ends at present, so the problem that the detection time hardware processing capability negotiated by a central processing unit at present cannot be reached to cause the a BFD conversation to be broken so as to cause frequent link jittering to lead to unnecessary system overhead can be avoided so as to improve the accuracy and resource utilization ratio of bidirectional forwarding detection.

The invention provides a processing and optimizing method and device for a dynamic language, equipment and a storage medium. The processing and optimizing method includes the steps: based on program codes in an intermediate file, obtaining an abstract syntax tree through analysis, obtaining type description information of the corresponding program codes through analysis based on the annotation information in the intermediate file, and adding the type description information to corresponding nodes on the abstract syntax tree; and traversing the abstract syntax tree so as to obtain attribute and type information owned by the object with the class structure based on the association relationship between the nodes and the type description information of the nodes, generating a hidden class, and attaching the hidden class to the class definition node corresponding to the object so as to generate an optimization code based on the abstract syntax tree with the hidden class. Therefore, by pre-generating the hidden class during compiling, the object layout information can be known in advance, so that the source file can be optimized based on the pre-generated hidden class to avoid the overhead of adding attributes to create the hidden class and expanding the object layout during running, and the running rate is increased.

The invention discloses an alarm association rule mining method. The method comprises the following steps: de-noising the current network alarm according to set field information and obtaining a weight parameter of the de-noised alarm according to an alarm characteristic attribute; carrying out window splitting on the de-noised alarm; constructing a transaction database, and constructing a weighted frequent mode tree by taking the de-noised alarm as a project, taking an alarm window as a transaction and combining the weight parameter of the de-noised alarm; obtaining a weighted frequent mode set according to the weighted frequent mode tree; and obtaining a weighted association rule for alarms in the weighted frequent mode set according to a relationship between the weighted frequent mode set and all the sub-sets thereof. According to the method disclosed in the invention, the efficient and correct mining of the alarm association rule can be realized. The invention furthermore discloses an alarm association rule mining device.

The embodiment of the invention provides a task dispatching method and a task dispatching device based on DSP (digital signal processing), wherein the method concretely comprises the following steps that task information occupying an exclusive lock is recorded in a data structure of the exclusive lock; when the P operation is executed on the exclusive lock in the current task, the task information occupying the exclusive lock is read from the data structure of the exclusive lock, in addition, the priority of the current task is compared with the priority of the task occupying the exclusive lock, and if the priority of the current task is higher than the priority of the task occupying the exclusive lock, the priority of the task occupying the exclusive lock is improved to the priority of the current task. The method and the device have the advantage that the real-time performance of the DSP is improved on the premise of not being limited by use scenes.

The present application provides a method and apparatus for Internet asset scanning discovery and service identification. The executing subject of the method is a computer connected to the Internet and provided with a scanning pool. The method includes: determining an idle node in the Internet; sending a disguised scanning data packet to an Internet asset corresponding to an IP address in the scanning pool, and buffering the IP address and port of the Internet asset locally; and then sending a session acknowledgement packet to the idle node; receiving a feedback packet returned by an idle nodein an asynchronous manner; determining an Internet asset corresponding to the feedback data packet according to a local buffer, and judging whether a port of the Internet asset is in an open state according to an identification number in the feedback data packet; if the port of the Internet asset is in an open state, the service provided by the Internet asset is identified according to a preset port and a service characteristic comparison relation library. The present application can effectively improve the efficiency of Internet asset scanning through the above means.