Data deduplication in a disaggregated storage system

    公开(公告)号:US12235811B2

    公开(公告)日:2025-02-25

    申请号:US17351733

    申请日:2021-06-18

    Abstract: A data deduplication process is performed in a storage system which includes storage nodes, and storage control nodes which can access data directly from each storage node. A first storage control node sends a message to a second storage control node to initiate a deduplication process with respect to a given data block and an original data block owned by the second storage control node. The second storage control node increments a reference counter associated with the original data block, and sends a message to the first storage control which includes metadata. The first storage control node uses the metadata to read the original data block from a given storage node, performs a data compare process to determine whether the given data block matches the original data block, and creates a reference to the original data block, if the given data block matches the original data block.

    Acceptance status classification of product-related data structures using models with multiple training periods

    公开(公告)号:US12229639B2

    公开(公告)日:2025-02-18

    申请号:US17078675

    申请日:2020-10-23

    Abstract: Techniques are provided for acceptance status classification of product-related data structures using models with multiple training periods. One method comprises obtaining data for a given product-related data structure; evaluating first features related to the given product-related data structure using the obtained data; applying the first features related to the given product-related data structure to one or more models trained using multiple different training periods to obtain a plurality of second features, wherein each of the second features indicates a prediction related to an acceptance status of the given product-related data structure by at least one model for a respective training period; and aggregating at least the second features to obtain a classification related to an aggregate acceptance status of the given product-related data structure. A weighting of at least some of the first and second features can be learned during a training phase.

    Facilitating per-CPU reference counting for multi-core systems with a long-lived reference

    公开(公告)号:US12229559B2

    公开(公告)日:2025-02-18

    申请号:US17504904

    申请日:2021-10-19

    Inventor: Anton Rang

    Abstract: Facilitating per-CPU reference counting for multi-core systems with a long-lived reference is provided herein. A system includes a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations. The operations include determining a first quantity of releases associated with an object in a data structure of the system and determining a second quantity of acquisitions associated with the object. The first quantity of releases can be distributed among respective first counters of processing elements of a group of processing elements. The second quantity of acquisitions can be distributed among respective second counters of the processing elements of the group of processing elements. Further, the operations can include, based on the second quantity of acquisitions and the first quantity of releases being determined to be a same value, implementing a removal of the object from the data structure.

    Performing workload migration in a virtualized system based on predicted resource distribution

    公开(公告)号:US12223363B2

    公开(公告)日:2025-02-11

    申请号:US17547944

    申请日:2021-12-10

    Abstract: Techniques for managing resources of a virtualized system involve acquiring historical distribution data about a virtualized system, the historical distribution data indicating a historical distribution of resources occupied by workloads on a plurality of host machines of the virtualized system over a predetermined historical time period. The techniques further involve generating predicted distribution data based on the historical distribution data, the predicted distribution data indicating an estimated distribution of resources occupied by the workloads on the plurality of host machines over a predetermined future time period. The techniques further involve performing workload migration at least once based on the predicted distribution data, the workload migration including migrating a workload of a first host machine whose first estimated quantity of occupied resources exceeds a high threshold to a second host machine whose second estimated quantity of occupied resources is below a low threshold.

    Remote replication with host encryption

    公开(公告)号:US12212656B2

    公开(公告)日:2025-01-28

    申请号:US17237749

    申请日:2021-04-22

    Abstract: Decrypting data at a first storage system that has been encrypted at a second, separate, storage system includes the first storage system requesting a key that decrypts the data from the second storage system, the second storage system determining if the first storage system is authorized for the key, the second storage system providing the key to the first storage system in response to the first storage system being authorized, a host that is coupled to the first storage system obtaining the key from the first storage system, and the host using the key to decrypt and access the data at the first storage system. The host and the first storage system may provide failover functionality for a system that includes the second storage system. The host may obtain the key from the first storage system in response to a failure of the system that includes the second storage system.

    Systems and methods for workload placement based on subgraph similarity

    公开(公告)号:US12204944B2

    公开(公告)日:2025-01-21

    申请号:US17505820

    申请日:2021-10-20

    Abstract: Techniques described herein relate to systems and methods for workload placement based on subgraph similarity. Such techniques may include obtaining an encoded workload graph based on receiving a workload execution request; using the encoded workload subgraph to obtain encoded graphs representing previous workload executions, encoded subgraphs representing infrastructures on which the workload were executed, resource usage information, and execution metrics; using the encoded infrastructure subgraphs using subgraph similarity to identify candidate infrastructure subgraphs, using an ML model to predict an execution metric for an execution of the workload using the candidate; and selecting a best candidate infrastructure on which to execute the workload based on the predicted execution results.

    Extending retention lock protection from on-premises to the cloud

    公开(公告)号:US12197392B2

    公开(公告)日:2025-01-14

    申请号:US18307575

    申请日:2023-04-26

    Abstract: Embodiments for retention locking a deduplicated file stored in cloud storage by defining object metadata for each object of the file, and comprising a lock count and a retention time based on an expiry date of the lock, with each object having segments, the object metadata further having a respective expiry date and lock count for each segment, where at least some segments are shared among two or more files. Also updating the lock count and retention time for all segments of the file being locked; and if the object is not already locked, locking the object using a retention lock defining a retention time and updating the object metadata with a new lock count and the retention time, otherwise incrementing the lock count and updating the retention time for the expiry date if expiry date of a previous lock is older than a current expiry date.

    Performance tuning a computer system in scaling domains based on quantified scalability

    公开(公告)号:US12190125B2

    公开(公告)日:2025-01-07

    申请号:US18105761

    申请日:2023-02-03

    Abstract: Systems and methods for performance tuning a computer system in scaling domains based on quantified scalability. A system includes a processor configured to: calculate an estimate of parallel fraction and speedup characteristic in a first domain D1 and in a second domain D2 for an application, the estimate being calculated using system performance measurements generated from previous processing iterations of one or more workloads of the application using a number, n, of cores in the first domain and a remaining number, N−n, of cores in the second domain to obtain performance values XD1(n) and XD2(N−n), wherein N represents a total number of cores; calculate the number of cores for the first domain using a quadratic equation generated from the parallel fraction and performance value in each domain; and execute the application in each domain using the number of cores for each domain.

    Method, electronic device, and computer program product for flushing data

    公开(公告)号:US12175113B2

    公开(公告)日:2024-12-24

    申请号:US17824317

    申请日:2022-05-25

    Abstract: In techniques for flushing data, based on a maturity level of a storage segment, the storage segment is inserted into a list to be flushed corresponding to the maturity level in a plurality of lists to be flushed, the plurality of lists to be flushed respectively correspond to different maturity levels, and the maturity level at least indicates a proportion of the number of data-written blocks to the total number of blocks of the storage segment; and the list to be flushed for the corresponding maturity level in the plurality of lists to be flushed is flushed to a disk array according to a descending order of the maturity levels. In this way, the bandwidth utilization of the disk array can be improved.

Patent Agency Ranking