Google Inc.

Walfredo Cirne

John Wilkes

David Oppenheimer

Brian K Grant

Joseph L Hellerstein

Cosmos Nicolaou

David P Bort

Todd Pu-Tse Wang

Jason Hickey

Kai-Peter Backman

Indranil Gupta

Todd Pu-tse Wang

Scott David Johnson

Rafal Sokolowski

Geeta Chaudhry

A method of specifying behavior among a group of computing tasks included in a request to be performed in a domain of computing resources is disclosed. Method steps include receiving, at a scheduler operably coupled to the domain, a p/f request, the received p/f request including a first group and a first relationship, the first group comprising at least a first p/f group element and a second p/f group element, the first relationship defining a desired behavior of the first and second p/f group elements with respect to each other during performance of the p/f request; determining whether the domain includes available computing resources capable of satisfying the first relationship; and in response to a determination that the domain includes available computing resources capable of satisfying the first relationship, allocating, with the scheduler, at least one available computing resource to fulfill the p/f request.

Specifying behavior among a group of computing tasks

Techniques and systems are disclosed for allocating resources between two or more resource pools in a computing resource environment. Allocation may be realized by identifying a first resource in the first resource pool; creating a second resource based on at least a portion of the identified first resource; adding the created second resource to the second resource pool; identifying at least a portion of the added second resource as unused with respect to the second resource pool; creating a third resource based on the identified portion of the second resource; and adding the created third resource to a resource pool other than the second resource pool.

Omega resource model: returned-resources

In a computing resource environment including at least two different resource name spaces, a method for generating a fully-qualified name for a resource based on a context-based name of that resource and a usage context is disclosed. Method steps include receiving the resource's name schema, its context-based name, and usage context; comparing an entry in the schema and an entry in the context-based name and identifying schema monikers missing from the context-based name. For a missing moniker, determining whether it is an attribute space (aspace) moniker. If it is an aspace moniker, adding it to a full name that includes the context-based name. If it is not an aspace moniker, determining the aspace associated with the missing moniker; searching the usage context for an association between the missing moniker, its aspace, and a value assigned to the missing moniker; and appending the searched-for items to the full name.

Omega names: name generation and derivation utilizing nested three or more attributes

Multiple scheduler verticals can allocate tasks to resources that are shared by the scheduler verticals. Information regarding a state of each resource may be stored in memory accessible by the multiple scheduler verticals, and a processor updates the information. The scheduler verticals schedule events to be performed by any of the resources, and submit updates to reflect the scheduled events in the information. In the event of conflicting events, an update corresponding to only one of the conflicting events is committed. Moreover, disruptions may be preplanned and scheduled so as to minimize impact on scheduled tasks.

Systems and methods for performing scheduling for a cluster

A system and method with the following features is disclosed: receiving a partial name of a resource having an attribute moniker and attribute value; receiving a context associated with an attribute space moniker, an attribute moniker and an attribute value; receiving a schema associated with a set that includes an attribute space moniker and an attribute moniker; identifying a missing moniker that is present in the set but not the partial name; adding a missing moniker that is an attribute space moniker to the partial name; when a missing moniker is not an attribute space moniker, identifying a context where the attribute moniker is associated with the missing moniker and an attribute space moniker is present in the partial name or set; adding the context's attribute moniker and attribute value to the partial name; and returning the partial name with the added monikers and attribute values as the full name.

Resource name generation and derivation utilizing attribute space monikers and their associated context

A method of verifying the validity of a fully-qualified resource name within an environment including at least two different resource name spaces. The method includes receiving a fully qualified resource name having two or more tokens; and reading a token from the name. If the token represents the start of a name space, adding the name space onto a stack; and reading another token from the name. If read token doesn't represent the start of a name space, determining if it represents an end of the name space. If it does represent the end of the name space, popping the top of the stack as the subsequent token from the name. If it doesn't represent the start or end of the name space, checking if the token is a valid moniker for the name space. If it is valid, reading another token from the name, and otherwise returning an error.

Omega names: name generation and derivation

Systems and methods are provided for generating backup tasks for a plurality of tasks scheduled to run in a computer cluster. Each scheduled task is associated with a target probability for execution, and is executable by a first cluster element and a second cluster element. The system classifies the scheduled tasks into groups based on resource requirements of each task. The system determines the number of backup tasks to be generated. The number of backup tasks is determined in a manner necessary to guarantee that the scheduled tasks satisfy the target probability for execution. The backup tasks are desirably identical for a given group. And each backup task can replace any scheduled task in the given group.

Growing community of inventors

Palo Alto, CA, United States of America

Walfredo Cirne