US 12,474,998 B2
Adaptive resource allocation using data criticality for protection policies
Avinash Kumar, Patna (IN); Mahesh Reddy Av, Bangalore (IN); and Terry O'Callaghan, Hopkinton, MA (US)
Assigned to Dell Products L.P., Round Rock, TX (US)
Filed by Dell Products L.P., Round Rock, TX (US)
Filed on Jul. 14, 2023, as Appl. No. 18/352,262.
Prior Publication US 2025/0021441 A1, Jan. 16, 2025
Int. Cl. G06F 11/14 (2006.01); G06N 20/00 (2019.01)
CPC G06F 11/1461 (2013.01) [G06F 11/1453 (2013.01); G06F 11/1464 (2013.01); G06N 20/00 (2019.01); G06F 2201/84 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method of adaptively allocating resources of a data protection system for application of a data protection policy on a data asset, comprising:
processing the data asset in a present execution of the policy;
analyzing a resource usage of the policy to produce an analyzed resource usage for resources comprising at least one of: CPU utilization, memory allocation, disk storage, and network bandwidth;
predicting, using the analyzed resource usage, a resource consumption by the policy on the data asset during the present execution of a data protection operation on a backup client storing the data asset to facilitate balancing resources used for the data asset to which the policy is applied;
quantifying the resource consumption as an adaptive resource value;
first classifying the backup client as critical or non-critical based on first attributes associated with the backup client;
first tagging the backup client with a client criticality tag;
second classifying the data asset of the first tagged backup client as critical or non-critical based on second attributes associated with the asset, wherein the first and second classifying each comprises a k-nearest neighbors (KNN) algorithm that determines the criticality of the data asset based on the first and second attributes;
second tagging the data asset with a data asset criticality tag after the first tagging to further tag the data asset with the data asset criticality tag along with the client criticality tag;
continuously providing, to a backup server of the data protection system, the predicted resource consumption to mark the data protection operation against all running policies including the policy;
modifying a resource usage of the policy to create an adapted policy based on the data asset criticality tag of the asset and the adaptive resource value to prioritize protection of data based on data criticality of both the data and backup clients to reduce a cost of performing the data protection policy, and to balance the resources used for the data asset; and
backing up the asset in accordance with the adapted policy.