81 results about "Converged storage" patented technology

The invention discloses a software-defined fusion storage method for a solid-state disc. An HFSL (host fusion storage layer) is added on a host, physical block characteristics of the solid-state disc and statistical information during usage are obtained directly, complex IO (input/output) scheduling and management of the solid-state disc are conducted in combination with upper-layer data access characteristics, and an address mapping method with variable allocation granularities is used for realizing persistent high performance of the solid-state disc. According to the method, performance optimization of an NAND flash and storage management fusing a PCM (phase change memory) and the NAND flash are realized on the host, and the outstanding performance of the PCM is used for making up performance defects of the NAND flash, so that the read/write performance of the solid-state disc is improved, and the service life of the solid-state disc is prolonged.

Method and device for processing network and storage related traffic is provided. The device includes a network port for interfacing with a network switch; a plurality of ports that receive a converged storage and network protocol related initialization request. Based on selectable criteria, a controller selects a port from among the plurality of ports to respond to the converged storage and network protocol related initialization request. The device also includes a translation module operationally coupled to the port for translating information between a converged network and storage protocol and a storage protocol; and a flexible port that is configured by a port controller to communicate with a storage system using the storage protocol or configured to operate as a network port communicating with a network device.

In an embodiment, a processor may be configured to detect a store exclusive operation followed by a memory barrier operation in a speculative instruction stream being executed by the processor. The processor may fuse the store exclusive operation and the memory barrier operation, creating a fused operation. The fused operation may be transmitted and globally ordered, and the processor may complete both the store exclusive operation and the memory barrier operation in response to the fused operation. As the fused operation progresses through the processor and one or more other components (e.g. caches in the cache hierarchy) to the ordering point in the system, the fused operation may push previous memory operations to effect the memory barrier operation. In some embodiments, the latency for completing the store exclusive operation and the subsequent data memory barrier operation may be reduced if the store exclusive operation is successful at the ordering point.

A fusion memory device and method that is capable of storing binary data in a compressed format and reading out the compressed binary data in a decompressed format is provided. The fusion memory device includes a main memory for storing data in a compressed format; a secondary memory for buffering binary data to be written within and read from the main memory; and a memory controller having a pair of compressor and decompressor. The memory controller controlling the secondary memory to buffer the binary data downloaded from a host device and the compressor compressing the binary data buffered with the secondary memory in a writing mode. The memory controller controlling the secondary memory to buffer the binary data read out from the main memory within the secondary memory and the decompressor decompressing the binary data buffered within the secondary memory in a reading mode.

The invention relates to a convergence method of an iSCSI (Internet Small Computer System Interface) and an FCP (Fiber Channel Protocol). The method realizes the convergence of the iSCSI and the FCP through a convergence storage module and comprises the following steps of: (1) initializing the convergence storage module, establishing the mapping between a logical address space and physical storage equipment in a data storage system and recognizing the type of the physical storage equipment; (2) parsing an SCSI command; and (3) converging the SCSI command; and (4) returning an execution result of the SCSI command. The invention supports the exchange of an iSCSI data packet and an FCP data packet on an SCSI layer, realizes the visualization of a heterogeneous network storage system and effectively improves the deployment flexibility and the expandability of a disaster recovery system.

The invention discloses a converged storage method of service source data. The method comprises the following steps of acquiring the service source data; defining a storage mode of the document storage database-based service source data; defining a storage mode of the graphics database-based service source data; reading a service source log, and according to the service source log, creating and updating a document storage database; reading documents in the document storage database according to periods, inserting the documents into a graphics database, and creating a source graph spectrum; according to the document storage database, updating the graphics database. According to the converged storage method of the service source data provided by the invention, balance between a multi-type search requirement and search efficiency is realized; the transparency is kept, the using experience of a user is improved, and uniform analysis and search operation are facilitated through a uniform search analyzer.

The invention discloses an efficient organization and management method of massive spatio-temporal data. The overall steps of the method are as follows: step 1, uploading data; 2, data processing; 3,data storage; 4, dynamically dividing the spatial data into blocks; a, constructing a vector data Key-Value data model; B. HDFS vector data being stored in blocks. The invention aims at the information access safety in the large concurrent access, high concurrency access, multi-source heterogeneous spatio-temporal information fusion storage and updating, performs research on cloud-based solutions,to realize the effective integrated management of massive heterogeneous data and improve the distribution, access and operation performance of spatial data in the distributed environment, at the sametime, the technology of GIS data distributed dynamic partition is studied to optimize the distribution efficiency of GIS service engine, and to meet the smart city geospatial data efficient storage management and sharing distribution.

The invention discloses an ontology fusion preprocessing method for multi-source heterogeneous resources. The method comprises the following steps: S1, extracting an entity concept in an ontology model; S2, extracting a relation concept in the ontology model; S3, extracting an attribute concept in the ontology model; S4, exporting a JSON data file; S5, discovering an obvious structure of the datato be acquired, and expressing the two structures by using a regular expression; s6, returning subscripts of the content meeting the regular expression structure in the source data; s7, acquiring sub-strings meeting conditions; s8, further performing character string matching on the sub-character strings; s9, fusing the labels of the source data; and S10, storing the data, wherein the storage format is a structured format. According to the method, the ontology modeling result can be converted into the JSON narration, and the syntax structure definition for consistent fusion of heterogeneous (semi-) structured data is given.

The invention provides an enterprise-level heterogeneous fusion memory management system. The enterprise-level heterogeneous fusion memory management system comprises a memory device management module, a memory pool management module, an advanced telecom computing architecture (ATCA) management module and a user management module. The memory device management module collects basic information and sends the collected basic information to the memory pool management module, the ATCA management module, and the user management module simultaneously. The memory pool management module manages and displays the information in a sorted mode and sends the information to the ATCA management module and the user management module. The ATCA management module analyzes and processes the received information comprehensively and sends the information after comprehensive analysis and process to the user management module. A fusion architecture with two layers of structures is adopted, and the fusion architecture is compatible with network storage, so fusion of various protocols are achieved. A bottom layer memory device is virtualized, and transparent memory service is provided to application, so maintenance of memory management can be carried out conveniently. The whole memory network is divided into a plurality of areas, each area forms a system automatically, and memory sources in the bottom layer can meet application requirement of local memory.

The invention discloses a resource monitoring method, device and terminal based on a hyper-converged storage system, relates to the technical field of data processing and is mainly intended to use nodes in the hyper-converged storage system to perform integrated management on monitoring data and to provide improved stability and efficiency for results output in data monitoring. According to the main technical scheme, at least one first node server in the hyper-converged storage system is used to acquire original monitoring data; the original monitoring data and corresponding merge informationare stored in a system monitoring database, wherein the system monitoring database is used for storing the monitoring data and the corresponding merge information; the monitoring data in the system monitoring database are merged based on the merge information, and merge results are stored in a corresponding monitoring data table. The resource monitoring method, device and terminal based on the hyper-converged storage system are mainly used to monitor resources of the hyper-converged storage system.

The invention discloses a data read-write control method and system based on a hyper-converged storage system and a terminal, relating to the technical field of data storage. Therefore, effective utilization of processing resources in the hyper-converged storage system is increased. The method comprises the following steps: searching corresponding virtual volume controllers in local nodes according to read-write requests when receiving the read-write requests at the local nodes, wherein the virtual volume controllers are used for controlling read-write operation of virtual volumes corresponding to virtual machines created in the hyper-converged storage system; sending requests for creating the virtual volume controllers in the local nodes to a manager if there is not any virtual volume controllers, wherein the manager is used for managing virtual volume controllers corresponding to all virtual volumes in the hyper-converged storage system; and enabling the virtual volume controllers tocontrol corresponding virtual volumes to perform corresponding read-write operation according to information fed back to the requests by the manager.

One embodiment facilitates a high-density converged storage system. During operation, the system receives, by a volatile memory of a storage device via a memory bus, data to be stored in a non-volatile memory of the same storage device. The system writes, by a controller of the storage device, the data directly to the non-volatile memory via the volatile memory, e.g., without using a serial bus protocol.

The invention belongs to the technical field of multi-core processor design, in particular to an off-chip accelerator applicable to a fusion memory of a 2.5D (2.5 dimensional) multi-core system. The off-chip accelerator comprises a logic control circuit (comprising an instruction decoder and a state register), a memory (organized in an array manner), an accelerator and a configurable internetwork. The logic control circuit receives a configuration packet from a processor end, configures a corresponding link via analysis, and decides whether output data of the memory returns to a processor via acceleration logic. The accelerator is applicable to the specific application oriented 2.5D multi-core processing system, and can reduce performance loss due to longer inter-chip access delay, the output data of the memory can be calculated and output by selecting different acceleration units, and the accelerator has certain flexibility.

The embodiment of the invention provides a fusion memory, a storage system and a deep learning calculation method. The fusion memory comprises an acceleration storage controller and a storage chip; the acceleration storage controller comprises a processor, a data path module and an acceleration processing module; the processor is used for scheduling the data path module according to the received storage command and storing externally input data to the storage chip corresponding to the storage command, and the data comprises to-be-calculated data; the processor is used for scheduling a data path module according to the obtained calculation control instruction and a loading mode corresponding to the calculation control instruction, and loading the to-be-calculated data to an acceleration processing module; the data path module is used for providing a transmission path for stored data and providing a transmission path for loading to-be-calculated data; the acceleration processing module is used for performing deep learning calculation on the to-be-calculated data through hardware acceleration; and the storage chip is used for storing data. Through the scheme, the calculation efficiency of deep learning calculation can be improved.

The invention discloses a remote heterogeneous multi-fusion storage method. According to the method, a process of cluster monitoring nodes is stored, a data structure of a storage architecture is maintained, a client acquires specific storage nodes for communication and read-write operation by calculating addressing after acquiring the data structure, and addressing rules are realized through a controllable, extensible and distributed data placement algorithm. The rules are configurable, wherein when configuration is performed at a geographic position level, remote storage can be realized, andhigh availability of a storage cluster is guaranteed; when configuration is performed from the perspective of equipment type, heterogeneous storage can be realized; and according to data use scene selection, a storage device with better performance is selected for data with high use frequency to improve read-write performance, and a device with slightly poor performance is selected for data withlow use frequency to lower use cost. The method can support multiple storage use modes such as object storage, block storage and file system storage at the same time, and therefore the remote heterogeneous multi-fusion storage method is formed.