Abstract: A method is implemented by a network device to configure the operation of a Peripheral Component Interconnect Express (PCIe) switch to enable an efficient transition from a first active processor in a first root complex to a backup processor in a second root complex. The method involves determining the first active processor in the first root complex and a set of backup processors and a set of root complexes, and configuring each root complex for independent PCIe switch communication. The method further includes detecting a failure of the active processor in the first root complex, selecting and notifying the backup processor and the second root complex to transition to be a second active processor and second root complex, and starting communication with PCIe devices using previously configured independent PCIe switch communication for the second processor of the second root complex.

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a system that includes a synthesis subsystem that automatically creates components for inclusion in a creative. The synthesis subsystem includes one or more modules that receive, as input, crawled content or metadata known or identified by the content serving system, and produce, as an output, a synthesized component for inclusion in a prospective creative. The system further includes an evaluation subsystem that evaluates quality of each of the synthesized components generated by the synthesis subsystem to determine measurable quality as against a baseline quality level.

Abstract: A method and apparatus for an in-database sharded queue that supports JMS session ordering is provided. Messages from a plurality of enqueuers are stored in a plurality of shards of a particular queue. For each enqueuer, all messages are stored in only one of the shards. Each shard includes one or more subshards, each subshard having one or more partitions of a queue table maintained by a relational database system. All messages from a first enqueuer are stored in partitions of the queue table that are assigned to a first shard of the plurality of shards. All messages from a second enqueuer are stored in partitions of the queue table that are assigned to a second shard of the plurality of shards. Subscribers of the particular queue are caused to dequeue messages from the plurality of shards.

Abstract: A system for transmission of packetized data in a voice activated computer network environment. The system can identify candidate canonicalized digital components that relate to secondary search locations. For each of the candidate canonicalized digital components the system can calculate transmissions scores that are used to select one of the candidate canonicalized digital components for transmission.

Abstract: Methods and systems for improved processor efficiency via reductions in repeated calculations are provided. A plurality of candidate sitelinks are identified in response to a search for online content. Each sitelink has associated with it a plurality of candidate creatives with which the sitelink may be presented to the user. The creatives are canonicalized to form clusters of candidate creatives. The sitelinks are also canonicalized. The creatives are matched to the candidate canonicalized sitelinks so as to provide enhanced sitelinks having increased relevance to the user search.

Abstract: Systems and methods for providing navigation filters within content items distributed via a computer network are provided. A navigation filter includes a heading and a plurality of items associated with the heading and can function as a summary of a resource associated with a distributed content item. A described method includes receiving a request for content from a user device, selecting a content item in response to the request, identifying one or more navigation filters associated with the selected content item, generating display data including the selected content item and the one or more navigation filters, and providing the display data for presentation to the user device.

Abstract: A method and apparatus for updating an application on a group of nodes is presented. According to one embodiment, an application is updated at a first node. The first node updates a registry to indicate that an update was performed at the first node and propagates the update to the registry to one or more second nodes. At a second node, the second node determines that one or more application updates are available at the first node. Upon such a determination, the second node requests one or more update packages from the first node. Based on an update policy associated with the second node, the second node updates the application using the one or more update packages.

Abstract: Techniques are disclosed for identifying “free” compute nodes that have not been configured as a member of a secondary storage system. As disclosed, a node may automatically detect that it is a free node (i.e., one available to join a distributed computing cluster which provides the secondary storage system or to form a new cluster). When a node detects that it is not part of a cluster, the node may build an initial network configuration and begin advertising itself as a free node. Once the initial network state is available, the node may broadcast advertisements observed by a management tool and joined to the secondary storage system.

Abstract: Methods, systems and apparatus, including computer programs encoded on a computer storage medium, for determining similar queries for image searches. In one aspect, a method includes generating, for each of a plurality of search queries, a selection vector for the search query, each selection vector including a plurality of vector elements, each vector element corresponding to a unique image, and the value of each vector element being proportional to a number of selections of its corresponding unique image in response to the unique image being presented as a search result for the search query. The method further includes selecting a first search query and a second search query from the plurality of search queries and determining, from the selection vectors for the first and second search queries, a similarity measure that is a measurement of the similarity of the first search query to the second query.

Abstract: A system, method, and computer program product are provided for conditionally sending a request for data to a node based on a determination. In operation, a first request for data is sent to a cache of a first node. Additionally, it is determined whether the first request can be satisfied within the first node, where the determining includes at least one of determining a type of the first request and determining a state of the data in the cache. Furthermore, a second request for the data is conditionally sent to a second node, based on the determination.

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for providing content. A search query is received. Search results responsive to the query are identified, including identifying a first search result in a top set of search results that is associated with a brand. Based at least in part on the query, one or more eligible content items are identified for delivery along with the search results responsive to the query. A determination is made as to when at least one of the eligible content items is associated with a same brand as the brand associated with the first search result. The first search result and one of the determined at least one eligible content items are combined into a combined content item and providing the combined content item as a search result responsive to the request.

Abstract: Methods and systems for submitting media object queries and receiving suggested answers for the media object queries. In one aspect, a method includes receiving from a first user a first media object and a first query relating to content in the first media object, presenting the first media object and the first query to multiple second users, receiving a suggested answer to the first query from each of two or more second users of the multiple second users, where at least two of the suggested answers are distinct, ranking the suggested answers, and presenting one or more of the ranked suggested answers to the first user.

Abstract: An example method for scaling event processing in a network environment is provided and includes maintaining a first portion of a decision tree at a first server in the network environment, delegating a second portion of the decision tree from the first server to a second server, processing event data substantially simultaneously at the first server using the first portion of the decision tree and at the second server using the second portion of the decision tree, wherein the processing comprises determining a match between the event data and information stored at nodes in the decision tree. In various embodiments, the decision tree is distributed across a plurality of servers in the network, wherein each participating server maintains a local copy of a respective portion of the decision tree and processes the event data using the respective portion.

Abstract: Systems and methods for detecting device or carrier change conversions based on device information and carrier information that is received with requests to provide content to communication devices is provided. A system can receive requests to provide content including a device identifier, an account identifier identifying an account agnostic to the device, and a carrier identifier. The system can serve content items in response to the requests to provide content that are related to communication devices different from the device associated with the device identifier or carriers different from the carrier associated with the carrier identifier. The system can detect, from the contents of a subset of the requests to provide content, that the account identifier is associated with a different device identifier or carrier identifier and identify one or more content items corresponding to the different device identifier or carrier identifier.

Abstract: Systems and methods for generating navigation filters are provided. A target resource is parsed and occurrences of data entries representing archetypal headings and/or archetypal items for a navigation filter are identified in the content of the target resource. A path to each of the identified occurrences is determined and used to construct a query for additional navigation filter headings and/or navigation filter items that have the same path as an identified occurrence. Identified occurrences of archetypal data entries in the target resource and the additional content having the same path as an identified occurrence are used as potential navigation filter headings and/or potential navigation filter items. Navigation filters are generated by associating one or more of the potential navigation filter items with one of the potential navigation filter headings.

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for providing content. A search query is received. Search results responsive to the query are identified, including identifying a first search result in a top set of search results that is associated with a brand. Based at least in part on the query, one or more eligible content items are identified for delivery along with the search results responsive to the query. A determination is made as to when at least one of the eligible content items is associated with a same brand as the brand associated with the first search result. The first search result and one of the determined at least one eligible content items are combined into a combined content item and providing the combined content item as a search result responsive to the request.

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for providing content. Search results responsive to a query are identified including a first search result in a top set of search results, the first search result associated with a first entity. A first eligible content item is identified for presentation along with the search results, the first eligible content item associated with the first entity. A combined content item is identified that is a combination of the first search result and first eligible content item and is to be presented as a search result responsive to the query. The combined content item is augmented including: identifying entities related to the first entity, identifying content items that are associated with the related entities, selecting at least one identified content item, and using content from the selected content items to augment the combined content item.

Abstract: Techniques are disclosed for identifying “free” compute nodes that have not been configured as a member of a secondary storage system. As disclosed, a node may automatically detect that it is a free node (i.e., one available to join a distributed computing cluster which provides the secondary storage system or to form a new cluster). When a node detects that it is not part of a cluster, the node may build an initial network configuration and begin advertising itself as a free node. Once the initial network state is available, the node may broadcast advertisements observed by a management tool and joined to the secondary storage system.

Abstract: Methods and systems for submitting media object queries and receiving suggested answers for the media object queries. In one aspect, a method includes receiving from a first user a first media object and a first query relating to content in the first media object, presenting the first media object and the first query to multiple second users, receiving a suggested answer to the first query from each of two or more second users of the multiple second users, where at least two of the suggested answers are distinct, ranking the suggested answers, and presenting one or more of the ranked suggested answers to the first user.