Microsoft Technology Licensing, LLC

Marwan E Jubran

Aleksandr Mikhailovich Gershaft

Vitalii Tsybulnyk

Weiping Hu

Ulrich Mueller

Vladimir Petrenko

Christopher Warner McCarron

Maksim Libenson

Sanjay Ramanujan

Vipins Gopinadhan

Igor Avramovic

Jiwu Liu

Saurabh Agarwal

Aritra DattaGupta

Luke Rafael Rodriguez

Muhammad Khizar Qazi

Andrew Tianze Wang

Willy Tanimihardja

Xiaoguang Fan

Clifford Thomas Dibble

Arka Dasgupta

Damien S Jose

Christopher W Mccarron

Methods, media, and systems for secure provisioning of servers within a cloud computing environment are provided for herein. In some embodiments, a management service can delegate provisioning of a server of the cloud computing environment to an imaging service. In response, the imaging service can generate an operating system image for the server and can utilize disk encryption to protect to operating system image. In embodiments, a volume encryption key of the disk encryption can be encrypted utilizing a public key of a trusted platform manager of the server, to produce an encrypted volume encryption key that is protected by the trusted platform module of the server. The encrypted operating system image and the encrypted volume encryption key can then be transmitted to the server to cause the server to be provisioned with the operating system image. Other embodiments may be described and/or claimed herein.

Secure provisioning of operating systems

Various methods and systems for implementing dependency management for an infrastructure are provided. A dependency service manager of the dependency management system includes a plurality of dependency management system components that facilitate automatically discovering and analyzing data to generate dependency data. The dependency data includes dependencies and dependents of dependency service tenants which are tenant services of the infrastructure. The dependency data is generated based on data retrieved by a plurality of collectors. A collector from the plurality of collectors is an agent that retrieves data associated with dependency service tenants based on a corresponding collection time attribute of the collector. A collector accesses data for generating the dependency data at one of the following collection times: design time, deployment time and runtime. The dependency data that is generated is then stored in a data store and communicated to a dependency service interface for user access to the dependency data.

Distributed computing dependency management system

A system and method for blocking path detection is provided. A job comprises tasks with at least some of the tasks dependent on other task(s). Each task is assigned to an ownership team. At a particular point in execution of the job, states of tasks of the job are identified. The states include one of three mutually exclusive states: waiting for another task/finished, in progress, and, blocked. When all the tasks with identified states of in progress or blocked are assigned to a particular ownership team, the particular ownership team is identified as on a blocking path. An action can be performed regarding the blocking path, for example, selected in accordance with policy-defined response actions such as generating an incident, escalating an existing incident, and/or sending a notification (e.g., an accumulated time on the blocking path can be calculated for each team with team(s) notified when certain threshold(s) are exceeded).

System and method for blocking path detection

An automated root-cause analysis (RCA) system may provide a fully automated platform that provides dependency and execution order modeling for tasks included in a capacity provisioning process, anomaly detection, ticket correlation, root-cause analysis, monitoring and feedback, and data visualization. The automated RCA system may continuously collect and store data for use in determining a root cause of a blockage on a capacity provisioning process. The blockage may be identified in a ticket generated by a cloud-computing system. The automated RCA system may receive the ticket and attempt to determine the root cause of the blockage based on root causes associated with previous tickets generated by the cloud-computing system. The automated RCA system may identify a true root cause, recommend repair items based on the true root cause, identify one or more responsible teams to drive a fix, and provide an estimated time for completion.

Automatic root cause analysis and prediction for a large dynamic process execution system

To improve the reliability of nodes that are utilized by a cloud computing provider, information about the entire lifecycle of nodes can be collected and used to predict when nodes are likely to experience failures based at least in part on early lifecycle errors. In one aspect, a plurality of failure issues experienced by a plurality of production nodes in a cloud computing system during a pre-production phase can be identified. A subset of the plurality of failure issues can be selected based at least in part on correlation with service outages for the plurality of production nodes during a production phase. A comparison can be performed between the subset of the plurality of failure issues and a set of failure issues experienced by a pre-production node during the pre-production phase. A risk score for the pre-production node can be calculated based at least in part on the comparison.

Node health prediction based on failure issues experienced prior to deployment in a cloud computing system

A computer-implemented method for managing a release of a software product includes obtaining a request for the release, the request including workflow action parameter data to define a release pipeline involving a plurality of software engineering systems configured to process data indicative of the software product, and executing, with a processor, a workflow to implement the release pipeline in accordance with the workflow action parameter data. Executing the workflow includes sending a series of instructions to the plurality of software engineering systems. A successive instruction in the series of instructions is sent based on whether a gating rule for the release is met.

Software release workflow management

In embodiments, methods and systems for implementing an infrastructure management system that supports hardware failure remediation are provided. A determination is made that a failure of a hardware component has occurred. The hardware component is part of a hardware composite. A remediation attribute of the hardware composite is accessed. The remediation attribute indicates a minimum operational bar for the hardware composite. The minimum operational bar is based on health metrics or optional and required components of the hardware composite. The bar supports determining whether the hardware composite should operate in a degraded state. A determination is made that the hardware composite if operating without the hardware component that has failed meets the minimum operational bar for the hardware composite. Operation of the hardware composite in the degraded state is initiated. In embodiments, hardware managers are associated with degraded state configurations to facilitate operating the hardware composite in the degraded state.

Infrastructure management system for hardware failure

A method for nonintrusive network load generation may include determining available resources in a distributed computing system, where the distributed computing system includes a plurality of computing devices and a target deployment. Based on an amount of available resources between the target deployment and a plurality of source computing devices, the plurality of source computing devices may be selected to generate a network load directed from the plurality of source computing devices to the target deployment. The plurality of source computing devices may be a subset of the plurality of computing devices in the distributed computing system. A network-traffic generator service may be provided to the plurality of source computing devices in order to generate the network load directed from the plurality of source computing devices to the target deployment. The performance of the distributed computing system in response to the generated network load may be monitored.

Nonintrusive dynamically-scalable network load generation

In various embodiments, methods and systems for modeling or representing hardware inventory based on a topology definition is provided. A topology definition having physical elements and logical elements corresponding to a physical topology layer and a logical topology layer is received. An instance definition for a rack, cluster, or datacenter for a hardware management operation is received. The instance identifies a physical topology and a logical topology for the operation. A hardware management service is executed, based on the topology definition and the instance definition, to manage the hardware inventory of the topology definition. Hardware management of the hardware inventory corresponding to the separately defined physical elements and logical elements is performed with attributes that are defined in the schema. The topology definition can include precision levels that indicate a hierarchical classification of topology components classified at a device level, a rack level, a cluster level, and a datacenter level.

Datacenter topology definition schema

Described herein is a system and method for dependency version conflict auto-resolution for executing job(s). During execution of a particular version of a workflow comprising tasks, information regarding a particular task to be executed is received. The particular task is dependent on another task. Information regarding the dependency is retrieved from a global dependency data structure (e.g., graph) that stores current dependency information regarding the tasks. When it is determined that the conflict exists with respect to the dependency of the particular task on the another task, information regarding the dependency and workflows is retrieved, information regarding a dependency change history regarding at least one of the plurality of tasks, information regarding the particular version of the workflow and a different version of the workflow. A resolution to the conflict can be identified based, at least in part, upon the retrieved information, and, a correct action performed.

Dependency version conflict auto-resolution

Embodiments of the present invention provide systems, methods, and computer storage media for automated standalone bootstrapping of hardware inventory that includes a provisioning workflow of discovering, validating, and configuring hardware inventory. In an exemplary embodiment, an in-rack computing device, such as a chassis manager or selected blade, initializes an intra-rack communication network with a hardware inventory of the rack. The in-rack computing device initiates intra-rack communication with the hardware inventory in the rack, using the intra-rack communication network, to discover at least a set of computing units in a hardware inventory. Intra-rack communications that include traffic received from the set of computing units. The in-rack computing device validates operational configurations of the hardware inventory based on comparing expected operational configuration parameters with actual operational configuration parameters, at least a portion of the operational configuration parameters retrieved from the information carried within the traffic received from the set of computing units.

Growing community of inventors

Redmond, WA, United States of America

Marwan E Jubran