Granted Patent B2

US 12,353,923 · App. 17/722,147 · Granted Jul 8, 2025

Inventors: William Jeffery White (Plano, TX); Said Tabet (Austin, TX)

Assignee: Dell Products L.P.

G06F9/5083G06F9/4881

Abstract

Claims (75)

1. A method for managing a distributed multi-tiered computing (DMC) environment, comprising:

identifying, by a global controller, a load management prediction event;

in response to identifying the load management prediction event:

obtaining offline data from local controllers associated with a plurality of DMC domains of the DMC environment;

generating load management predictions using the offline data;

adjusting local controller scheduling policies based on the load management predictions;

adjusting global controller scheduling based on the load management predictions;

making a first determination that the load management predictions require infrastructure adjustments;

in response to the first determination:

initiating infrastructure capacity adjustment based on the load management predictions

making a second determination, based on the load management predictions and adjusted local controller scheduling policies, that a future scheduling load associated with a set of local controller scheduler instances exceeds a threshold; and

in response to the second determination:

initiating an adjustment of a quantity of the set of local controller scheduler instances to satisfy the future scheduling load.

2. The method of claim 1 , wherein initiating the adjustment of local controller scheduler instances comprises increasing the quantity of the set of local controller scheduler instances.

3. The method of claim 1 , wherein the load management predictions comprise:

short-term load management predictions;

medium-term load management predictions; and

long-term load management predictions.

4. The method of claim 1 , wherein the load management predictions are generated, at least in part, using a metaheuristic algorithm.

5. The method of claim 1 , wherein the load management predictions comprise:

global controller load management predictions;

local controller load management predictions;

domain load management predictions; and

application load management predictions.

6. The method of claim 1 , wherein the plurality of DMC domains comprises:

edge domains, wherein an edge domain of the edge domains comprises an edge domain device set;

core domains, wherein a core domain of the core domain comprises a core domain device set; and

cloud domains, wherein a cloud domain of the cloud domains comprises a cloud domain device set.

7. The method of claim 1 , wherein adjusting local controller scheduling policies based on the load management predictions comprises selecting scheduling policy priorities for the local controllers.

8. The method of claim 7 , wherein the scheduling policy priorities specify which scheduling algorithms are used to schedule application tasks.

9. 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 a distributed multi-tiered computing (DMC) environment, comprising:

identifying, by a global controller, a load management prediction event;

in response to identifying the load management prediction event:

obtaining offline data from local controllers associated with a plurality of DMC domains of the DMC environment;

generating load management predictions using the offline data;

adjusting local controller scheduling policies based on the load management predictions;

adjusting global controller scheduling based on the load management predictions;

making a first determination that the load management predictions require infrastructure adjustments;

in response to the first determination:

initiating infrastructure capacity adjustment based on the load management predictions;

making a second determination, based on the load management predictions and adjusted local controller scheduling policies, that a future scheduling load associated with a set of local controller scheduler instances exceeds a threshold; and

in response to the second determination:

initiating an adjustment of a quantity of the set of local controller scheduler instances to satisfy the future scheduling load.

10. The non-transitory computer readable medium of claim 9 , wherein initiating the adjustment of local controller scheduler instances comprises increasing the quantity of the set of local controller scheduler instances.

11. The non-transitory computer readable medium of claim 9 , wherein the load management predictions comprise:

short-term load management predictions;

medium-term load management predictions; and

long-term load management predictions.

12. The non-transitory computer readable medium of claim 9 , wherein the load management predictions are generated, at least in part, using a metaheuristic algorithm.

13. The non-transitory computer readable medium of claim 9 , wherein the load management predictions comprise:

global controller load management predictions;

local controller load management predictions;

domain load management predictions; and

application load management predictions.

14. The non-transitory computer readable medium of claim 9 , wherein the plurality of DMC domains comprises:

edge domains, wherein an edge domain of the edge domains comprises an edge domain device set;

core domains, wherein a core domain of the core domain comprises a core domain device set; and

cloud domains, wherein a cloud domain of the cloud domains comprises a cloud domain device set.

15. The non-transitory computer readable medium of claim 9 , wherein adjusting local controller scheduling policies based on the load management predictions comprises selecting scheduling policy priorities for the local controllers.

16. The non-transitory computer readable medium of claim 15 , wherein the scheduling policy priorities specify which scheduling algorithms are used to schedule application tasks.

17. A system for managing a distributed multi-tiered computing (DMC) environment, the system comprising:

a DMC environment; and

a global controller of the DMC environment, comprising a processor and memory, and configured to:

identify a load management prediction event;

in response to identifying the load management prediction event:

obtain offline data from local controllers associated with a plurality of DM domains of the DMC environment;

generate load management predictions using the offline data;

adjust local controller scheduling policies based on the load management predictions;

adjust global controller scheduling based on the load management predictions;

make a first determination that the load management predictions require infrastructure adjustments;

in response to the first determination:

initiate infrastructure capacity adjustment based on the load management predictions;

make a second determination, based on the load management predictions and adjusted local controller scheduling policies, that a future scheduling load associated with a set of local controller scheduler instances exceeds a threshold; and

in response to the second determination:

initiate an adjustment of a quantity of the set of local controller scheduler instances to satisfy the future scheduling load.

Assignments (1)

Continuity (1)

References Cited (73)