SYSTEM AND METHOD FOR OPTIMIZING RESOURCE UTILIZATION IN A CLUSTURED OR CLOUD ENVIRONMENT

• US 20140115163A1
• Filed: 01/22/2013
• Published: 04/24/2014
• Est. Priority Date: 10/24/2012
• Status: Active Grant

• Current Assignee: MessageOne, Inc. (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Matousek, Jon Franklin
• Granted Patent: US 9,491,114 B2
• Time in Patent Office: Days
• US Class Current: 709/226
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: H04L 47/762 triggered by the network
• CPC Class Codes: H04L 47/788 Autonomous allocation of resources
• CPC Class Codes: H04L 47/822 Collecting or measuring resource availability data

Abstract

In one embodiment, a method includes calculating ideal resource apportionments from a current set of consumable resources for each of a plurality of reservations. Each reservation corresponds to one of a plurality of customers. The method further includes running an apportionment process relative to the plurality of reservations. The running includes attempting to apportion to each reservation its ideal resource apportionment. The running yields an actual resource apportionment for each reservation. Each customer'"'"'s actual resource apportionment comprises a sum of the actual resource apportionments for the customer'"'"'s reservations. In addition, the method includes, responsive to an indication of unapportioned resources following the running, performing a first optimization to increase resource utilization by at least one needy customer.

Claims

1. A method comprising:

on a computer cluster comprising a plurality of computers, calculating ideal resource apportionments from a current set of consumable resources for each of a plurality of reservations;

wherein each reservation corresponds to one of a plurality of customers;

wherein each customer'"'"'s ideal resource apportionment comprises a sum of the ideal resource apportionments for the customer'"'"'s reservations;

running an apportionment process relative to the plurality of reservations, the running comprising attempting to apportion to each reservation its ideal resource apportionment;

wherein the running yields an actual resource apportionment for each reservation;

wherein each customer'"'"'s actual resource apportionment comprises a sum of the actual resource apportionments for the customer'"'"'s reservations;

responsive to an indication of unapportioned resources following the running, performing a first optimization to increase resource utilization by at least one needy customer; and

wherein the at least one needy customer comprises one or more customers whose actual resource apportionments are less than their ideal resource apportionments.

2. The method of claim 1, wherein performing the first optimization comprises:

identifying the at least one needy customer;

identifying reservations of the at least one needy customer that can utilize at least a portion of the remaining resources;

grossing up the ideal resource apportionment for each identified reservation according to the at least one needy customer'"'"'s ideal resource apportionment; and

re-running the apportionment process relative to the identified reservations, the re-running comprising attempting to apportion to each identified reservation its grossed-up ideal resource apportionment.

3. The method of claim 1, comprising, responsive to an indication of unapportioned resources following the first optimization, performing a second optimization to increase resource utilization across all customers.

4. The method of claim 3, wherein performing the second optimization comprises:

identifying reservations of any customer that can utilize at least a portion of the remaining resources;

grossing up the ideal resource apportionments for the identified reservations; and

re-running the apportionment process relative to the identified reservations, the re-running comprising attempting to exhaust the remaining resources.

5. The method of claim 1, wherein the apportionment process comprises:

identifying one or more provisions of the current set that has remaining available resources;

wherein each provision comprises resources of the current set that serve a same set of resource profiles;

for each provision of the one or more provisions;

generating a set of available resource profiles for the provision;

acquiring at least one sorted set, the at least one sorted set comprising a plurality of profile entries;

filtering the at least one sorted set by the available resource profiles to yield at least one filtered sorted set; and

for each profile entry of the at least one filtered sorted set;

fetching the profile entry;

fetching a sorted profile set corresponding to the fetched profile entry;

computing a resource quantity to apply to each reservation of the sorted profile set; and

apportioning the resource quantity to each reservation of the sorted profile set.

6. The method of claim 5, wherein the acquiring comprises:

creating a set of reservations, the set comprising reservations that have not yet attained at least one of their ideal resource apportionments and their grossed-up ideal resource apportionments;

generating a sorted profile set for each resource profile represented in the set of reservations;

placing each reservation of the set of reservations into an appropriate sorted profile set based on the reservation'"'"'s resource profile;

computing a smallest total need for each sorted profile set; and

generating the at least one sorted set, the at least one sorted set comprising a profile entry for each sorted profile set, each profile entry comprising the smallest total need for the profile set.

7. The method of claim 5, the method comprising repeating the apportionment process responsive to at least one provision having remaining available resources and an existence of at least one needy reservation that can utilize at least a portion of the remaining available resources.

8. The method of claim 5, wherein, for each reservation, computing a resource quantity to apply comprises:

computing a maximum quantity of resources to apply from the provision;

accessing a smallest total need for the sorted profile set;

determining a minimum of the maximum quantity and the smallest total need;

normalizing the determined minimum per reservation;

calculating a maximum resource quantity that the reservation can accept without exceeding at least one of its ideal resource apportionment and its grossed-up ideal resource apportionment; and

wherein the resource quantity to apply to the reservation comprises a minimum of the normalized minimum and the calculated maximum for the reservation.

9. The method of claim 5, wherein apportioning the resource quantity to each reservation comprises, for each reservation:

deducting the reservation'"'"'s need by the resource quantity;

deducting a need for a customer to which the reservation corresponds by the resource quantity; and

deducting an applicable provision'"'"'s remaining resources by the resource quantity.

10. The method of claim 1, wherein each actual resource apportionment comprises a flow-control clocking weight.

11. The method of claim 10, comprising:

maintaining a priority queue of the plurality of reservations;

responsive to a determination that at least one consumable resource in the current set is free;

identifying a first reservation in the priority queue;

performing a suitability function on the first reservation;

responsive to a determination that the at least one consumable resource has been deemed suitable for the first reservation, assigning the at least one consumable resource to the first reservation; and

responsive to a determination that the at least one consumable resource has not been deemed suitable for the first reservation, identifying a next reservation in the priority queue.

12. An information handling system, comprising:

a computer cluster comprising a plurality of computers, wherein the computer cluster is operable to;

calculate ideal resource apportionments from a current set of consumable resources for each of a plurality of reservations;

wherein each reservation corresponds to one of a plurality of customers;

wherein each customer'"'"'s ideal resource apportionment comprises a sum of the ideal resource apportionments for the customer'"'"'s reservations;

run an apportionment process relative to the plurality of reservations, the running comprising attempting to apportion to each reservation its ideal resource apportionment;

wherein the run yields an actual resource apportionment for each reservation;

wherein each customer'"'"'s actual resource apportionment comprises a sum of the actual resource apportionments for the customer'"'"'s reservations;

responsive to an indication of unapportioned resources following the running, perform a first optimization to increase resource utilization by at least one needy customer; and

wherein the at least one needy customer comprises one or more customers whose actual resource apportionments are less than their ideal resource apportionments.

13. The information handling system of claim 12, wherein operability to perform the first optimization comprises operability to:

identify the at least one needy customer;

identify reservations of the at least one needy customer that can utilize at least a portion of the remaining resources;

gross up the ideal resource apportionment for each identified reservation according to the at least one needy customer'"'"'s ideal resource apportionment; and

re-run the apportionment process relative to the identified reservations, the re-running comprising attempting to apportion to each identified reservation its grossed-up ideal resource apportionment.

14. The information handling system of claim 12, wherein the computer cluster is operable to:

responsive to an indication of unapportioned resources following the first optimization, perform a second optimization to increase resource utilization across all customers.

15. The information handling system of claim 14, wherein operability to perform the second optimization comprises operability to:

identify reservations of any customer that can utilize at least a portion of the remaining resources;

gross up the ideal resource apportionments for the identified reservations; and

re-run the apportionment process relative to the identified reservations, the re-running comprising attempting to exhaust the remaining resources.

16. The information handling system of claim 12, wherein the apportionment process comprises:

identifying one or more provisions of the current set that has remaining available resources;

wherein each provision comprises resources of the current set that serve a same set of resource profiles;

for each provision of the one or more provisions;

generating a set of available resource profiles for the provision;

acquiring at least one sorted set, the at least one sorted set comprising a plurality of profile entries;

filtering the at least one sorted set by the available resource profiles to yield at least one filtered sorted set; and

for each profile entry of the at least one filtered sorted set;

fetching the profile entry;

fetching a sorted profile set corresponding to the fetched profile entry;

computing a resource quantity to apply to each reservation of the sorted profile set; and

apportioning the resource quantity to each reservation of the sorted profile set.

17. The information handling system of claim 16, wherein the operability to acquire comprises operability to:

create a set of reservations, the set comprising reservations that have not yet attained at least one of their ideal resource apportionments and their grossed-up ideal resource apportionments;

generate a sorted profile set for each resource profile represented in the set of reservations;

place each reservation of the set of reservations into an appropriate sorted profile set based on the reservation'"'"'s resource profile;

compute a smallest total need for each sorted profile set; and

generate the at least one sorted set, the at least one sorted set comprising a profile entry for each sorted profile set, each profile entry comprising the smallest total need for the profile set.

18. The information handling system of claim 16, wherein, for each reservation, operability to compute a resource quantity to apply comprises operability to:

compute a maximum quantity of resources to apply from the provision;

access a smallest total need for the sorted profile set;

determine a minimum of the maximum quantity and the smallest total need;

normalize the determined minimum per reservation;

calculate a maximum resource quantity that the reservation can accept without exceeding at least one of its ideal resource apportionment and its grossed-up ideal resource apportionment; and

wherein the resource quantity to apply to the reservation comprises a minimum of the normalized minimum and the calculated maximum for the reservation.

19. The information handling system of claim 16, wherein operability to apportion the resource quantity to each reservation comprises, for each reservation, operability to:

deduct the reservation'"'"'s need by the resource quantity;

deduct a need for a customer to which the reservation corresponds by the resource quantity; and

deduct an applicable provision'"'"'s remaining resources by the resource quantity.

20. The information handling system of claim 12, comprising:

maintaining a priority queue of the plurality of reservations;

responsive to a determination that at least one consumable resource in the current set is free;

identifying a first reservation in the priority queue;

performing a suitability function on the first reservation;

responsive to a determination that the at least one consumable resource has been deemed suitable for the first reservation, assigning the at least one consumable resource to the first reservation; and

responsive to a determination that the at least one consumable resource has not been deemed suitable for the first reservation, identifying a next reservation in the priority queue.

21. A computer-program product comprising a computer-usable medium having computer-readable program code embodied therein, the computer-readable program code adapted to be executed to implement a method comprising:

calculating ideal resource apportionments from a current set of consumable resources for each of a plurality of reservations;

wherein each reservation corresponds to one of a plurality of customers;

wherein each customer'"'"'s ideal resource apportionment comprises a sum of the ideal resource apportionments for the customer'"'"'s reservations;

running an apportionment process relative to the plurality of reservations, the running comprising attempting to apportion to each reservation its ideal resource apportionment;

wherein the running yields an actual resource apportionment for each reservation;

wherein each customer'"'"'s actual resource apportionment comprises a sum of the actual resource apportionments for the customer'"'"'s reservations;

responsive to an indication of unapportioned resources following the running, performing a first optimization to increase resource utilization by at least one needy customer; and

wherein the at least one needy customer comprises one or more customers whose actual resource apportionments are less than their ideal resource apportionments.