Techniques for analytics-driven hybrid concurrency control in clouds
AssigneeTelefonaktiebolaget LM Ericsson (publ)
InventorsEugen FELLER, Julien FORGEAT, Muhammad Mudassar NAZAR
Life of this patent
Estimating the term…
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.
Claims21
A method in a hybrid resource allocation module implemented by one or more computing devices, the method comprising:
- receiving, from a first application of a plurality of applications, a first request for resource availability information of resources provided by a set of computing devices;
- determining, based at least in part on a use of a machine learning model, that the first application is categorized into an optimistic cluster of applications, wherein the machine learning model categorizes applications into the optimistic cluster or a pessimistic cluster;
- providing the resource availability information to the first application;
- and
- setting a lock to prevent any application that is categorized into the pessimistic cluster from being able to obtain resource availability information while the lock is set, wherein any application that is categorized into the optimistic cluster can still obtain available resource information while the lock is set.
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 implement a hybrid resource allocation module to perform operations comprising:
- receiving, from a first application of a plurality of applications, a first request for resource availability information of resources provided by a set of computing devices;
- determining, based at least in part on a use of a machine learning model, that the first application is categorized into an optimistic cluster of applications, wherein the machine learning model categorizes applications into the optimistic cluster or a pessimistic cluster;
- providing the resource availability information to the first application;
- and
- setting a lock to prevent any application that is categorized into the pessimistic cluster from being able to obtain resource availability information while the lock is set, wherein any application that is categorized into the optimistic cluster can still obtain available resource information while the lock is set.
A system comprising:
- one or more computing devices providing resources as part of a cloud computing environment;
- and
- a hybrid resource allocation module implemented by another one or more computing devices, the hybrid resource allocation module comprising instructions which, when executed by the one or more computing devices, cause the hybrid resource allocation module to:
- receive, from ones of a plurality of applications, requests for resource availability information of the resources provided by the one or more computing devices;
- determine, based at least in part on a use of a machine learning model, for an application that provided a request whether the application is categorized into an optimistic cluster of applications or into a pessimistic cluster of applications;
- and
- when the requesting application is in the optimistic cluster, provide the resource availability information to the requesting application and ensure that a lock is set to prevent any application that is categorized into the pessimistic cluster from being able to obtain resource availability information while the lock is set, wherein any application that is categorized into the optimistic cluster can still obtain available resource information while the lock is set.
Dependencies are inferred from the claim text.