65 results about "Replication (computing)" patented technology

A computing system performs real-time replication of database management system transactions. The computing system includes a source DBMS, a replication service and a data lakehouse. The source DBMS stores at least one source transaction table recording at least one source transaction and generates at least one recovery log indicating at least one modification in the at least one source transaction table. The replication service replicates the at least one source transaction recorded in the at least one source transaction table by generating at least one data file having a first data format. The data lakehouse stores at least one lakehouse table corresponding to the at least one source transaction table and having a second data format different from the first data format, and to modify the at least one lakehouse table based on the at least one data file.

Systems and methods include an Information Handling System (IHS) that is adapted to diagnose root causes of issues reported by hardware and / or software of the IHS. Telemetry is monitored that specifies operating status information for hardware components of the IHS. Stress event are detected that are related to the hardware components. One or more stress tests are identified for evaluation of each hardware component that is related to stress event. While monitoring the telemetry, the stress tests are conducted in order to replicate the detected stress event. When the detected stress event is replicated, a root cause hardware component of the IHS is determined based on machine learning evaluation of the telemetry generated during replication of the stress event.

In some examples, a first computing system receives a first metadata change made to metadata of a first instance of an object stored at a storage location of the first computing device. The first computing system associates, with the first metadata change, a time of the first metadata change. Based on receiving replication of a second instance of the object from a second computing system, the first computing system determines that a second metadata change was made to metadata of the second instance of the object. Further, the second instance of the object includes a time of the second metadata change. The first computing system at least one of: resolves a conflict between the first metadata change and the second metadata change based at least on comparing the times, or merges the first metadata change with the second metadata change into the metadata for the object.

Methods, systems, and devices for data management are described. A first computing node within a first set of computing nodes may receive, from a network gateway, a first replication request that is associated with a first replication job for the first cluster of computing nodes to provide copies of a first set of one or more computing snapshots to a second cluster of computing nodes. The first set of computing nodes may determine that a second computing node different from the first computing node will service the data replication request. A network bandwidth allocation may be increased for the first computing node that received the replication request directly from the network gateway while the network bandwidth allocation for the second computing node is maintained for the second computing node based on the second computing node receiving the request via an internal redirection of the request.

Systems and methods for providing selective data-replication among nodes of a distributed multi-tenancy MSSP architecture for performing secure orchestration and automated response (SOAR) are provided. According to one embodiment a master SOAR node of an MSSP receives multiple messages via a secure router coupling a computing environment of the MSSP in communication with respective computing environments of multiple customers of the MSSP. The messages contain information regarding alerts relating to network infrastructure of the customers and the information is controlled by data sharing policies implemented by tenant SOAR nodes within the respective computing environments of the customers. Based on an investigation into an alert relating to a network infrastructure of a particular customer, the master SOAR node causes a workflow to be remotely executed by a tenant SOAR node within the computing environment of the particular customer.

Verifying that data has been correctly replicated to a replication target, including: replicating a dataset stored at a first computing system to a second computing system; and determining, based at least on a comparison of a first hash and a second hash, validity of the dataset stored at the second computing system, wherein the first hash is generated by applying a hash function to a copy of the dataset that is stored at the first computing system and the second hash is generated by applying the hash function to a copy of the dataset that is stored at the second computing system.

A power state framework executes a workload in supported computing device types and configurations during a sequence of power state transitions to simulate user experiences. Workloads can be combined, e.g., default workload plus add-on workload(s). Device states are monitored, collected, reported, and analyzed to identify faults, which are prioritized for repair, e.g., based on the projected impact of the faults on users, e.g., if the users were to experience the faults. Computing device states include states of attached peripheral devices and wireless or wired connectivity before and after power state transitions. Device state indicates device stability throughout supported device power state transitions. Devices under test can be on debuggers during execution of workload and power state transitions. Automated, accelerated simulation of the execution of end user workloads and power state transitions in actual computing devices saves time and avoids replication errors compared to manual user-driven device testing.

The invention discloses a switching method and system of a redundant server, and solves the problems of detection delay and data inconsistency in traditional switching through cooperative work of hardware layer redundancy (redundant components, multi-path architecture and power supply protection), intelligent fault detection (multi-dimensional monitoring and second-level heartbeat) and dynamic load balancing and data consistency guarantee (synchronous replication and seamless switching technology). Experiments show that the scheme shortens the annual downtime of the system, improves the availability, is suitable for key business scenes such as financial transactions and cloud computing, and ensures the service continuity and the data reliability.

The invention relates to the technical field of communication, in particular to a resource management method and device, equipment and a medium, and the method comprises the steps: obtaining a task sequence hardware computing chain resource request, used for indicating a request demand, of application software; matching a hardware computing link corresponding to the task sequence from the mapping relation according to the hardware computing link resource request; returning an operation handle of the hardware calculation link to the application software; after a data processing request obtained by writing to-be-processed data into the operation handle by application software is received, processing the to-be-processed data in sequence according to the hardware calculation link; application software only needs to copy the to-be-processed data once, so that data copying from a large amount of memory to endpoint equipment is omitted; when the data is processed on the endpoint device, the link information of the hardware calculation link is carried, and the data can be output to the next endpoint device according to the calculation link information after calculation is completed, so that a large amount of path occupation and memory replication are reduced, and the utilization rate of various resources is improved.

According to some embodiments, a digital core for a cloud computing environment is associated with an enterprise (e.g., the digital core may comprise enterprise resource planning software optimized for an in-memory database). Moreover, an integration inspector framework may establish a checkpoint between an application system (e.g., a source application system to provide information via the digital core and / or a target application system to receive information via the digital core) and the digital core. The integration inspector framework then automatically validates a communication arrangement associated with the application system and performs a data replication test, in connection with the application system and the digital core, using test data. A result of the data replication test is then evaluated and indicated (e.g., to a user via a user interface).

Disclosed techniques and solutions can provide improved snapshot replication. Typically, an initial replica obtained using snapshot replication is periodically updated. However, the update process can unnecessarily consume computing resources if data in a source data object has not changed with respect to data in a replica data object. Disclosed techniques check to determine whether a snapshot replica is out of date before obtaining a new snapshot. The checks can be performed on manual request or on the occurrence of triggers, such as receiving a query that accesses the replica data object or according to a schedule. Information for current and prior versions of the remote data object can be compared to determine whether a replica is out of date, such as digest values of contents of the remote data object or timestamps associated with the remote data object.

A power state framework executes a workload in supported computing device types and configurations during a sequence of power state transitions to simulate user experiences. Workloads can be combined, e.g., default workload plus add-on workload(s). Device states are monitored, collected, reported, and analyzed to identify faults, which are prioritized for repair, e.g., based on the projected impact of the faults on users, e.g., if the users were to experience the faults. Computing device states include states of attached peripheral devices and wireless or wired connectivity before and after power state transitions. Device state indicates device stability throughout supported device power state transitions. Devices under test can be on debuggers during execution of workload and power state transitions. Automated, accelerated simulation of the execution of end user workloads and power state transitions in actual computing devices saves time and avoids replication errors compared to manual user-driven device testing.

The application discloses a method for automatically discovering block devices and copying changed data in real time and a storage medium, belongs to the technical field of cloud computing, and aims to solve the technical problem of how to realize automatic discovery of block devices and real-time copying of data of newly added disks, reduce the invasiveness to user servers, and increase the continuity of user services. The technical scheme is as follows: the method is to load a block device data real-time copying kernel module into a linux kernel, register block device information, poll and scan or actively trigger by listening to block device events to query newly mounted / dismounted disks, enable / disable the real-time copying function of the corresponding disks, and simultaneously, in the case of not interrupting read-write I / O, real-time copy the disks in use.

Disclosure is made of methods, apparatus and system for migrating virtual machines (VMs) between source and destination in a computing environment and, more specifically, to replication based migration. VMs migration is controlled so as to manage transferal of data associated with one or more VMs from a source location to a destination location to meet certain user definable or system constraints. Dynamic control and adjustment of system parameters associated with the migration is also disclosed.

A disclosed computing device capable of achieving high data rates in link diagnostic testing by replicating traffic may include (1) a packet forwarding device configured to generate, as part of a link diagnostic test, traffic that falls below a certain threshold by a deficit and (2) a packet replication device configured to (A) receive at least a portion of the traffic from the packet forwarding device and (B) replicate the portion of traffic to generate supplemental traffic that compensates for the deficit, and (3) an interface configured to transmit the traffic and the supplemental traffic to a remote device via a link as part of the link diagnostic test. Various other apparatuses, systems, and methods are also disclosed.

Embodiments of the present application disclose a metadata backup recovery method and device, electronic equipment and a storage medium, wherein the method is applied to a computing node in a distributed database management system, comprising: starting the computing node, the computing node storing a metadata table, used for recording a replication position of the computing node in a DDL log through coordinate data; pulling the DDL log from a metadata node according to the coordinate data, executing a DDL command in the DDL log to update local metadata of the computing node, and updating the coordinate data of the metadata table, so that the recovery and reconstruction of the distributed database computing node can be conveniently and accurately realized by using the local transaction processing capability of the computing node.

A computer can connect with a middleware computing system. The middleware computing system may use application programming interfaces (APIs), webhooks, file-based integration, database replication, message queues, websockets, or direct integration to establish connections with different computing devices. The computer may request a data structure classification for an external data structure stored in a remote computing device. The middleware computing system can receive the request, identify the connection that the middleware computing device has with the remote computing device, and retrieve records for transactions performed by the external data structure from the remote computing device. The middleware computing system can use metadata in the records to automatically determine a data structure type of the data structure. The middleware computing system can generate instructions that cause the computer to link the external data structure with the profile.

The application belongs to the technical field of artificial intelligence, graph data processing and high-performance computing, and discloses a communication-efficient memory fusion method and system for super-memory graph neural network training and a medium. A task-specific memory management scheme, zero-copy CPU-GPU transmission, a layout-aware GPU intercommunication pipeline and an out-neighbor gradient aggregation mechanism are used to significantly reduce the overall data communication overhead while ensuring training accuracy. In the fusion memory-aware graph partitioning, a same subgraph is maintained by the storage layout of multiple GPUs. For P task partitioning characteristics and T task partitioning vertex dimensions, the number of graph partitions required by the system is reduced, thereby reducing neighbor replication, and the overall CPU-GPU communication overhead of the system is successfully reduced by 50% to 64%. The back propagation of the out-neighbor gradient aggregation of the application changes the back propagation into an out-neighbor aggregation process in the preprocessing stage.

A computing system performs real-time replication of database management system transactions. The computing system includes a source DBMS, a replication service and a data lakehouse. The source DBMS stores at least one source transaction table recording at least one source transaction and generates at least one recovery log indicating at least one modification in the at least one source transaction table. The replication service replicates the at least one source transaction recorded in the at least one source transaction table by generating at least one data file having a first data format. The data lakehouse stores at least one lakehouse table corresponding to the at least one source transaction table and having a second data format different from the first data format, and to modify the at least one lakehouse table based on the at least one data file.

A computing device (101), a method, and a computer program product for replicating a three-dimensional label mask (20) of a three-dimensional volumetric data set (190) in a network of computing devices (100), wherein the device (101) receives, from a further computing device (102) of the network, a set of segmentation input data (21) describing an edit (2-1) of the label mask (20) at the further computing device (102), the set of segmentation input data comprising current position data, size data, and a label identifier associated with a spherical segmentation object (102-1) used by the second computing device (102) to interactively edit the label mask. The replication module (101-10) replicates an edited label mask of a further computing device (102) associated with a set of segmentation input data by calculating respective ellipsoids corresponding to the spherical segmentation object (102-1) stretched inversely proportional to the shapes of the voxels in the label mask, determining voxels (vi) of the label mask having positions inside the segmentation object based on the ratio of the distance between each voxel position in all three dimensions and the center position of the ellipsoid divided by the respective semi-axis lengths of the ellipsoid, the division being performed by negating the exponents of each denominator using respective bitwise operations on a graphics processing unit of the computing device (101), and updating the determined voxels in the replicated label mask (10) according to the label identifiers of the segmentation input data (21).

Methods, systems, and devices for data management are described. A first computing node within a first set of computing nodes may receive, from a network gateway, a first replication request that is associated with a first replication job for the first cluster of computing nodes to provide copies of a first set of one or more computing snapshots to a second cluster of computing nodes. The first set of computing nodes may determine that a second computing node different from the first computing node will service the data replication request. A network bandwidth allocation may be increased for the first computing node that received the replication request directly from the network gateway while the network bandwidth allocation for the second computing node is maintained for the second computing node based on the second computing node receiving the request via an internal redirection of the request.

A high-throughput data transformation and upserting system for distributed computing environments, comprising: a data ingestion module configured to receive structured, semi-structured, and unstructured data from one or more distributed data sources, including databases, cloud platforms, enterprise systems, streaming infrastructures, and external computing devices; a transformation module configured to convert the received data into predefined processing formats through data mapping, normalization, filtering, aggregation, and schema matching operations performed across multiple distributed computing nodes; and an upsertion module configured to insert new records and update existing records in one or more distributed databases while maintaining synchronization and consistency of the records across distributed computing resources.a synchronization module configured to coordinate the management of distributed transactions, replication activities, and consistency operations between interconnected databases and compute nodes; a workload management module configured to distribute processing workloads across distributed compute resources to improve throughput efficiency, scalability, and resource utilization for data-intensive processing operations; and a monitoring module configured to detect duplicate records, synchronization conflicts, operational failures, processing delays, and data inconsistencies related to distributed transformation and upsertion activities.

The invention discloses a data synchronization method, a database system and a computing device cluster, and belongs to the technical field of databases. The database system comprises a master region and m slave regions. The master region includes a master replication cluster and a master database node. The master replication cluster includes replication nodes that are compute nodes that are logically different from the master database nodes. And after the main replication cluster obtains the to-be-synchronized data generated by the main database node, generating at least one replication task corresponding to the to-be-synchronized data. And the main replication cluster adds the at least one replication task to a task queue. And the master replication cluster concurrently obtains the at least one replication task from the task queue, and concurrently synchronizes the data to be synchronized to the m slave regions according to the at least one replication task. The problem that the to-be-synchronized data cannot be synchronized to the slave region in time due to resource limitation of the master database node can be avoided.

Creating a replica of a storage system, including: receiving, by a first storage system from a computing device, data to be stored on the first storage system; reducing, by the first storage system, the data using one or more data reduction techniques; sending, from the first storage system to the second storage system, the reduced data, wherein the reduced data is encrypted; and sending, from the second storage system to a third storage system, the reduced data, wherein the reduced data is encrypted.

The present application discloses an expert-parallel processing method and system, an electronic device, and a storage medium, which solve the problem of load imbalance on computing cards that carry expert execution, thereby improving the processing efficiency of the computing cards. In the expert-parallel processing system, W experts are deployed from a host node to N computing cards by means of expert replication, wherein each of the N computing cards comprises at least two experts having different expert weights. In an embodiment of the present application, on the basis of text processing units respectively activated by the W experts, M text processing units are assigned to at least two experts on each of the N computing cards, so as to obtain text processing units respectively activated by the at least two experts on each of the N computing cards. The total number of text processing units activated by the at least two experts and corresponding to a same expert weight across different computing cards is equal to the number of text processing units corresponding to the same expert weight among the W experts obtained by routing partitioning.

A computer-implemented method, according to one approach, includes using a determined classification of a predetermined file to determine a data compliance ruleset that applies to data of the predetermined file. In response to a determination that the data of the predetermined file is scheduled to be moved and / or replicated from a source node to a target node of an edge computing environment, the determined data compliance ruleset is applied to the data of the predetermined file. The method further includes preventing the scheduled movement and / or replication of the data of the predetermined file from occurring, in response to a determination that the scheduled movement and / or replication of the data of the predetermined file violates at least one rule of the determined data compliance ruleset.

The application provides an automatic synthesis method of a convolutional neural network accelerator for in-memory computing, which can obtain a convolutional neural network weight replication candidate strategy under each parameter combination by using an analog annealing method based on a convolutional neural network inference task, a peak power consumption constraint of an in-memory computing accelerator and a hardware parameter look-up table of the in-memory computing accelerator, obtain a macro cell mapping scheme with optimal performance by using a genetic method, and obtain a convolutional neural network weight replication candidate strategy, a hardware parameter combination of the in-memory computing accelerator and a hardware resource allocation scheme corresponding to the scheme. The method provided by the application greatly reduces the professional threshold of architecture design, improves the efficiency of in-memory computing accelerator design, saves the cost of expert design, and effectively enriches the optimization strategy of accelerator architecture configuration scheme to guarantee the design quality of the in-memory accelerator configuration scheme.

Embodiments of the present invention involve composing synchronization structures that enable increased synchronization control. A virtual synchronization structure enables synchronization using structure traversal without replicating the full data set. This results in computing systems that are at a desired level of synchronization without full data replication. This method has the advantage of enabling limited sharing between systems and also speed.

A source computing system may transmit to a target computing system a replication request to replicate data. The replication request may comprise a checksum derived from a portion of the data to be replicated. The target system may use the checksum to obtain data from a local computing system corresponding to a matching checksum in a table corresponding to the local computing system. The target system may unicast a request to the local computing system based on a checksum associated with the local system in a master table matching the checksum received in the replication request. The target system may broadcast, to local computing systems, a request that comprises the checksum indicated in the replication request. The target system may construct the data to be replicated using data received from the local computing system(s), responsive to the unicast or broadcast, instead of replicating the data from the source computing system.

Techniques for generating replication data from a data source to be stored in a target location are described herein. A computing system can receive, from a client device, client requirements associated with a dataflow from the data source to the target location. The client requirements can include an expected data freshness value and an expected data query latency value. Additionally, the computing system can process the expected data freshness value with one or more machine-learned models to generate an extraction framework for extracting data from the data source. Moreover, the computing system can process the expected data query latency value with the one or more machine-learned models to generate a loading framework for loading data to the target location. Furthermore, the computing system can copy the replication data from the data source to the target location based on the extraction framework and the loading framework.