Microsoft Technology Licensing, LLC

Matthew Snider

Gopal Kakivaya

Nikola Puzovic

Yang Li

Hua-Jun Zeng

Lu Xun

Girish Mittur Venkataramanappa

Todd F Pfleiger

Ankur Jauhari

Anurag Gupta

Kumar Gaurav Khanna

Zhengguo Sun

Milan Micić

Drazen Sumic

Aleksa Brkic

Ivan Nedic

Jelena Petrovic

Neda M Pistinjat

Vesna Todorovic

Nikola Pavlovic

Maja Stikic

Ana Trajkovic

Manoj Singiresu

Abhiram Chivukula

Systems and methods related to prediction-based resource orchestration in a service fabric are described. An example method for reducing an interruption rate associated with a customer application offered as a collection of services includes predicting one or more quiet time periods associated with each of the service replicas associated with a service fabric, where each of the one or more quiet time periods corresponds to a low value of a predicted load specifying a consumption of a metric by a respective service replica on a given node. The method further includes during the one or more quiet time periods predicted by a trained machine learning model, performing an impact-less failover for one or more of the service replicas by scheduling a move for the one or more service replicas from a first node associated with the service fabric to a second node associated with the service fabric.

Prediction-based resource orchestration in a service fabric

A data processing system implements techniques for provisioning and operating databases including receiving a request to provision a new database that includes multiple replicas. The data processing system provisions the new database including the plurality of replicas. The plurality of replicas includes a primary replica, one or more secondary replicas, and one or more delta replicas. The primary replica is configured to handle read and write operations. The one or more secondary replicas include a copy of data and transactional log information of the primary replica and are unable to handle write operations. The one or more delta replicas are a copy of the transactional log information of the primary replica and are unable to handle read operations or write operations.

Techniques for stateful persisted service resource reduction

In various implementations, methods and systems resource balancing in a distributed computing environment are provided. A client defined resource metric is received that represents a resource of nodes of the cloud computing platform. A placement plan for job instances of service applications is generated. The placement plan includes one or more movements that are executable to achieve a target placement of the job instances on the nodes. It is determined that the placement plan complies with placement rules. Each placement rule dictates whether a given job instance of the job instances is suitable for placement on a given node of the nodes. The placement plan is executed based on determining that the target placement of the job instances improves balance of resources across the nodes of the cloud computing platform based on the resource represented by the client defined resource metric.

Balancing resources in distributed computing environments

A resource balancer for a cluster of nodes in a cloud computing system includes a load calculation module that receives load reports for respective load metrics associated with the nodes, wherein the load reports are indicative of respective loads on a plurality of resources provided by the nodes, calculates, based on the received load reports, an expected distribution of loads across the nodes for each load metric, and calculates, based on the expected distribution of loads, probabilities that each of the load metrics will exceed a respective predetermined capacity for each node. A load balancing module selectively balances, based on the calculated probabilities, the loads on the resources provided by the nodes.

Statistical resource balancing of constrained microservices in cloud PAAS environments

Growing community of inventors

Redmond, WA, United States of America

Matthew Snider