Techniques for analytics-driven hybrid concurrency control in clouds

Patent metadata

Field	Value
Publication number	US-10015062-B2
Application number	US-201514949703-A
Country	US
Kind code	B2
Filing date	Nov 23, 2015
Priority date	Nov 23, 2015
Publication date	Jul 3, 2018
Grant date	Jul 3, 2018

Abstract

Exemplary techniques for analytics-driven hybrid concurrency control in clouds are disclosed that include a hybrid resource allocation module that can concurrently utilize an optimistic allocation scheme alongside a pessimistic allocation scheme. Machine learning techniques utilizing previous activity history of applications can be used to train a cluster model that is integrated by a hybrid resource allocation module to classify applications in either a pessimistic cluster or an optimistic cluster that identifies under which scheme requests from the applications will be processed.

First claim

What is claimed is:

1. A method in a hybrid resource allocation module executing at a computing device for efficiently providing access to cloud resources to a plurality of applications, the method comprising:
receiving, at the hybrid resource allocation module from a first application of the plurality of applications, a request for available resource information of the cloud resources, wherein the first application is attempting to launch a task using at least some of the cloud resources;
determining, by the hybrid resource allocation module, whether the first application has been categorized into an optimistic cluster or a pessimistic cluster;
responsive to determining that the first application has been categorized into the optimistic cluster, setting a lock to prevent any of the plurality of applications categorized into the pessimistic cluster from being able to request any available resource information while the first application is still attempting to launch the task, wherein the lock does not prevent any other applications categorized into the optimistic cluster from being able to request available resource information while the first application is still attempting to launch the task; and
providing the available resource information to the first application.

2. The method of claim 1 , further comprising:
while the lock is set and while the first application is still attempting to launch the task,
receiving a second request for the available resource information from a second application of the plurality of applications, wherein the second application is attempting to launch a second task using at least some of the cloud resources,
determining whether the second application has been categorized into the optimistic cluster or the pessimistic cluster, and
responsive to determining that the second application has been categorized into the optimistic cluster, providing the available resource information to the second application.

3. The method of claim 2 , further comprising:
receiving, from the first application, a request to launch the task;
causing the task to be launched;
receiving, from the second application, a request to launch the second task; and
determining, by the hybrid resource allocation module, whether a resource conflict results from the request to launch the second task.

4. The method of claim 1 , further comprising:
while the lock is set and while the first application is still attempting to launch the task,
receiving a second request for the available resource information from a second application of the plurality of applications, wherein the second application is attempting to launch a second task using at least some of the cloud resources,
determining whether the second application has been categorized into the optimistic cluster or the pessimistic cluster, and
responsive to determining that the second application has been categorized into the pessimistic cluster, waiting to provide the available resource information to the second application until the lock is no longer set.

5. The method of claim 4 , further comprising:
releasing the lock;
setting a second lock to prevent all other applications categorized into the pessimistic cluster and also all of the applications categorized into the optimistic cluster from being able to request any available resource information while the second application is still attempting to launch the second task; and
providing the available resource information to the second application.

6. The method of claim 1 , further comprising generating a model based upon at least a history of activity involving the plurality of applications, wherein the model can be executed to determine which of the plurality of applications to categorize into the optimistic cluster and which of the plurality of applications to categorize into the pessimistic cluster.

7. The method of claim 6 , wherein said generating of the model is based upon at least:
one or more resource reservation request conflicts from the history; and
one or more task launch requests from the history.

8. The method of claim 6 , further comprising:
executing the model, resulting in at least one of the plurality of applications that was previously categorized into the pessimistic cluster to now be categorized into the optimistic cluster.

9. A non-transitory computer-readable storage medium having instructions which, when executed by one or more processors of a computing device, cause the computing device to efficiently provide access to cloud resources to a plurality of applications by performing operations comprising:
receiving, from a first application of the plurality of applications, a request for available resource information of the cloud resources, wherein the first application is attempting to launch a task using at least some of the cloud resources;
determining whether the first application has been categorized into an optimistic cluster or a pessimistic cluster;
responsive to determining that the first application has been categorized into the optimistic cluster, setting a lock to prevent any of the plurality of applications categorized into the pessimistic cluster from being able to request any available resource information while the first application is still attempting to launch the task, wherein the lock does not prevent any other applications categorized into the optimistic cluster from being able to request available resource information while the first application is still attempting to launch the task; and
providing the available resource information to the first application.

10. The non-transitory computer-readable storage medium of claim 9 , wherein the operations further comprise:
while the lock is set and while the first application is still attempting to launch the task,
receiving a second request for the available resource information from a second application of the plurality of applications, wherein the second application is attempting to launch a second task using at least some of the cloud resources,
determining whether the second application has been categorized into the optimistic cluster or the pessimistic cluster, and
responsive to determining that the second application has been categorized into the optimistic cluster, providing the available resource information to the second application.

11. The non-transitory computer-readable storage medium of claim 10 , wherein the operations further comprise:
receiving, from the first application, a request to launch the task;
causing the task to be launched;
receiving, from the second application, a request to launch the second task; and
determining whether a resource conflict results from the request to launch the second task.

12. The non-transitory computer-readable storage medium of claim 9 , wherein the operations further comprise:
while the lock is set and while the first application is still attempting to launch the task,
receiving a second request for the available resource information from a second application of the plurality of applications, wherein the second application is attempting to launch a second task using at least some of the cloud resources,
determining whether the second application has been categorized into the optimistic cluster or the pessimistic cluster, and
responsive to determining that the second application has been categorized into the pessimistic cluster, waiting to provide the available resource information to the second application until the lock is no longer set.

13. The non-transitory computer-readable storage medium of claim 12 , wherein the operations further comprise:
releasing the l

Assignees

• Ericsson Telefon Ab L M
• Ericsson Telefon Ab L M

Inventors

• Feller Eugen
• Forgeat Julien
• Nazar Muhammad Mudassar

Classifications

• H04L41/0894

  Policy-based network configuration management · CPC title

• G06F9/5072 Primary

  Grid computing · CPC title

• H04L41/5054 Primary

  Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components · CPC title

• H04L41/0893

  Assignment of logical groups to network elements · CPC title