Granted Patent B2

US 12,625,742 · App. 18/304,848 · Granted May 12, 2026

Inventors: Parminder Singh Sethi (Ludhiana, IN); Anay Kishore (Bangalore, IN)

Assignee: Dell Products L.P.

G06F9/5083G06F9/5072G06F2209/501G06F2209/5016G06F2209/502G06F2209/503G06F2209/505

Abstract

Claims (76)

1 . A method for managing loads in data clusters, comprising:

identifying, by a first load management module (LMM) of a first data cluster, a load performance decline event associated with the first data cluster;

in response to identifying the load performance decline event:

making a first determination that there are no active authenticated connections between the first LMM of the first data cluster and any other LMM executing on any other data cluster;

in response to the first determination:

accessing, from a database of the first data cluster, LMM information, wherein the LMM information comprises configuration parameters and connection bandwidths associated with other LMMs executing in other data clusters;

identifying a second LMM of a second data cluster using the LMM information, wherein identifying the second LMM comprises selecting the second LMM whose configuration parameters match a similarity threshold relative to configuration parameters of the first LMM;

converting current system time to coordinated universal time (UTC);

generating a secure string identifier using a current UTC time and a secure string parameter;

encrypting the secure string identifier and a static alphanumeric key;

establishing an authenticated connection using the encrypted secure string identifier and the static alphanumeric key;

sending requests to the second LMM over the authenticated connection using the secure string identifier, wherein the second LMM services the requests using the second data cluster;

making a second determination that load performance of the first data cluster is restored;

in response to the second determination:

setting the authenticated connection to standby; and

servicing second requests using the first data cluster.

2 . The method of claim 1 , wherein the secure string identifier is associated with an expiration timeframe specified by the secure string parameters.

3 . The method of claim 2 , wherein the method further comprises:

after setting the authenticated connection to standby:

identifying the expiration of the secure string identifier based on the expiration timeframe; and

terminating the authenticated connection.

4 . The method of claim 1 , wherein the load performance decline event comprises determining current data cluster performance metrics are below a baseline.

5 . The method of claim 1 , wherein the load performance decline event comprises identifying an occurrence of a predicted data cluster load performance decline window.

6 . A method for managing loads in data clusters, comprising:

identifying, by a first load management module (LMM) of a first data cluster, a load performance decline event associated with the first data cluster;

in response to identifying the load performance decline event:

selecting a second LMM of a second data cluster, wherein the second LMM is associated with an authenticated connection with the first LMM, wherein selecting the second LMM of the second data cluster comprises:

making a first determination that there are no active authenticated connections between the first LMM of the first data cluster and any other LMM executing on any other data cluster;

in response to the first determination:

accessing, from a database of the first data cluster, LMM information, wherein the LMM information comprises configuration parameters and connection bandwidths associated with other LMMs executing in other data clusters;

identifying the second LMM using the LMM information, wherein identifying the second LMM comprises selecting the second LMM whose configuration parameters match a similarity threshold relative to configuration parameters of the first LMM;

converting current system time to coordinated universal time (UTC);

generating a secure string identifier using a current UTC time and secure string parameters;

encrypting the secure string identifier and a static alphanumeric key;

establishing the authenticated connection using the encrypted secure string identifier and the static alphanumeric key;

sending requests to the second LMM using a secure string identifier associated with the authenticated connection, wherein the second LMM services the requests using the second data cluster;

making a second determination that load performance of the first data cluster is restored;

in response to the second determination:

setting the authenticated connection to standby; and

servicing second requests using the first data cluster.

7 . The method of claim 6 , wherein the secure string identifier is associated with an expiration timeframe specified by the secure string parameters.

8 . The method of claim 7 , wherein the method further comprises:

after sending the requests to the second LMM:

identifying the expiration of the secure string identifier based on the expiration timeframe; and

terminating the authenticated connection.

9 . The method of claim 6 , wherein the load performance decline event comprises identifying current data cluster performance metrics are below a baseline.

10 . The method of claim 6 , wherein the load performance decline event comprises identifying an occurrence of a predicted data cluster load performance decline window.

11 . The method of claim 10 , wherein the predicted data cluster load performance decline window is generated using:

a prediction model;

current data cluster performance metrics;

current request information; and

baseline data.

12 . The method of claim 11 , wherein the predicted data cluster load performance decline window specifies a predicted future timeframe that the performance metrics will be below baseline data.

13 . A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for managing loads in data clusters, the method comprising:

identifying, by a first load management module (LMM) of a first data cluster, a load performance decline event associated with the first data cluster;

in response to identifying the load performance decline event:

selecting a second LMM associated with a second data cluster, wherein the second LMM is associated with an authenticated connection with the first LMM, wherein selecting the second LMM of the second data cluster comprises:

making a first determination that there are no active authenticated connections between the first LMM of the first data cluster and any other LMM executing on any other data cluster;

in response to the first determination:

accessing, from a database of the first data cluster, LMM information, wherein the LMM information comprises configuration parameters and connection bandwidths associated with other LMMs executing in other data clusters;

identifying the second LMM using the LMM information, wherein identifying the second LMM comprises selecting the second LMM whose configuration parameters match a similarity threshold relative to configuration parameters of the first LMM;

converting current system time to coordinated universal time (UTC);

generating a secure string identifier using a current UTC time and secure string parameters;

encrypting the secure string identifier and a static alphanumeric key;

establishing the authenticated connection using the encrypted secure string identifier and the static alphanumeric key;

sending requests to the second LMM using a secure string identifier associated with the authenticated connection, wherein the second LMM services the requests using the second data cluster;

making a second determination that load performance of the first data cluster is restored;

in response to the second determination:

setting the authenticated connection to standby; and

servicing second requests using the first data cluster.

14 . The non-transitory computer readable medium of claim 13 , wherein the secure string identifier is associated with an expiration timeframe specified by the secure string parameters.

15 . The non-transitory computer readable medium of claim 14 , wherein the method further comprises:

after sending the requests to the second LMM:

identifying the expiration of the secure string identifier based on the expiration timeframe; and

terminating the authenticated connection.

16 . The non-transitory computer readable medium of claim 13 , wherein the load performance decline event comprises identifying current data cluster performance information below a baseline.

Assignments (1)

Continuity (1)

References Cited (29)