International Business Machines Corporation

Yahoo! Inc.

Lenovo (singapore) Pte. Ltd.

Excalibur Ip, LLC

Benjamin Clay Reed

Utkarsh Srivastava

Philip Lewis Bohannon

Flavio Junqueira

Andrew Stephen Tomkins

Lance Warren Russell

Shanmugasundaram Ravikumar

Mark Andrew Smith

Stephen P Morgan

Randal Chilton Burns

Daniel Norin Meredith

Edward Gustav Chron

Darrell D Long

Ted J Bonkenburg

Maysam Yabandeh

Ying Chen

Vanja Josifovski

Marcus Felipe Fontoura

Hovey Raymond Strong

Brian Frank Cooper

Steven R Welch

Christopher Olston

Chris Olston

Yang Zhang

Khaled Elmeleegy

Erwin Tam

Amruta Sadanand Joshi

Edward Robinson

Michael P Peercy

Described is a reliable and efficient design for a transactional status oracle ('ReTSO'), which supports large-scale storage systems. ReTSO uses a centralized scheme and implements snapshot isolation, a property that guarantees that read operations read a consistent snapshot of the data store. The centralized scheme of ReTSO enables a lock-free commit algorithm that prevents unreleased locks of a failed transaction from blocking other transactions. Each transaction submits the identifiers of modified rows to the ReTSO, the transaction is committed only if none of the modified rows is committed by a concurrent transaction, and a log of the committed transaction is cached. To recover from failures, before a client is notified of a change, the change cached is first persisted into a write-ahead log, such that the memory state can be fully reconstructed by reading from the write-ahead log. Furthermore, ReTSO implements a lazy, auto-garbage collector hash map that reduces the number of memory accesses per request.

Lock-free transactional support for large-scale storage systems

In a system for storing and retrieving a plurality of records, the plurality of records associated with a ledger, a client issues read and write requests associated with one of the plurality of records, a plurality of record servers responds to the requests received from the client, and a management server maintains and coordinates, between the client and the record servers, information associated with the ledger, records, and record servers.

System and method for logging operations

The present invention provides a method, system and computer program product for exposing at least one local service as at least one web service. In an exemplary embodiment, the method includes receiving a call for the at least one web service, where the call comprises a request and a response. The method also includes delegating the call to a client bundle. The method further includes obtaining the at least one local service from a service registry via the client bundle. The method also includes invoking at least one method on the at least one local service via the client bundle. The method further includes returning a response from the invoked method for the at least one web service.

Method and system of mapping at least one web service to at least one OSGi service and exposing at least one local service as at least one web service

In one embodiment, a first one of a plurality of message processors receives a request with respect to a topic, and determines whether the first message processor itself is currently responsible for the topic. If so, then the first message processor services the request. If not, then the first message processor determines whether any of the other ones of the message processors is currently responsible for the topic. If so, then the request is redirected to another message processor to be serviced. If not, then the first message processor selects one of the message processors to be responsible for the topic and to service the request.

Distributed publish/subscribe system

A method and apparatus for altering a page presenting search results is provided. The query dispatcher receives one or more query terms. Based on the query terms, the search engine generates a set of search results and advertisements. A parallel search dispatched by entertainment rating and selection engine generates a set of content items based on the one or more query terms and an additional one or more constraint terms. The entertainment item rating and selection engine selects a content item from the set of content items. The selection may be random, based on past user responses, or responses of users belonging to particular clusters. The entertainment item injector then replaces one of the search results or advertisements with the content item. The content item is presented to the user on a search results page. The content item contains a feedback mechanism to collect user responses. The entertainment item rating and selection engine then derives the quality of the entertainment item from the collected user responses.

Syndicating humor

The present invention provides a method and system of performing a current filesystem backup with respect to a previous filesystem backup using directorywise hardlinking for a computer filesystem. In an exemplary embodiment, the method and system include (1) identifying all largest unchanged subtrees in the previous filesystem backup with respect to a current computer filesystem and (2) creating directorywise hardlinks in the current filesystem backup to the all largest unchanged subtrees. In an exemplary embodiment, the creating includes for each of the largest unchanged subtrees, forming a first pointer from the current filesystem backup to the each of the largest unchanged subtrees. In a further embodiment, the creating includes for each of the largest unchanged subtrees, establishing a second pointer from the each of the largest unchanged subtrees to a directory in the current filesystem backup that corresponds to the first pointer.

Filesystem backup using directorywise hardlinking for a computer filesystem

Methods and systems are provided that may be utilized for spilling in query processing environments.

Method or system for spilling in query environments

The present invention provides a method and system of mapping. In an exemplary embodiment, the method and system include (1) creating a proxy bundle corresponding to the at least one web service, (2) registering a proxy web service corresponding to the proxy bundle in a service registry, (3) searching for the at least one web service in the service registry, (4) returning a reference to the proxy web service from the service registry in response to the searching, and (5) invoking at least one method on the returned reference, thereby invoking the at least one web service.

Method and system of mapping at least one web service to at least one OSGi service

Techniques that support trail-based exploration by a user of a repository of documents are described herein. In one embodiment, trail definition data that specifies a trail is received. The trail includes an ordered series of waypoints including a trailhead, intermediate waypoints, and one or more trailends. In some embodiments, deadends may also be defined in the trial. A particular waypoint in the ordered series of waypoints is established as a current waypoint. Search terms can be received from a user to cause a search to be performed. It is then determined whether the search satisfies matching criteria associated with a waypoint that immediately follows the current waypoint in the ordered series of waypoints. If so, the user advances to the next waypoint. Otherwise, the user remains at the current waypoint. Finally, if a trailend is reached, then an action such as rewarding the user in some way may be performed.

Trail-based exploration of a repository of documents

The present invention, in an example embodiment, provides a special-purpose formal language and translator for the parallel processing of large databases in a distributed system. The special-purpose language has features of both a declarative programming language and a procedural programming language and supports the co-grouping of tables, each with an arbitrary alignment function, and the specification of procedural operations to be performed on the resulting co-groups. The language's translator translates a program in the language into optimized structured calls to an application programming interface for implementations of functionality related to the parallel processing of tasks over a distributed system. In an example embodiment, the application programming interface includes interfaces for MapReduce functionality, whose implementations are supplemented by the embodiment.

Formal language and translator for parallel processing of data

A system and method are provided to accessing a set of web pages within a context. The set of web pages may be represented using a graph comprising a set of nodes and a set of edges. First and second probability distributions may be generated over a set of nodes of the graph using a first algorithm to indicate a measure of closeness among the set of web pages. A characterization may be determined with respect to context for a web page from a second subset of web pages based on the first and second probability distributions. An indication of the characterization of the web page from the second subset of web pages may be outputted.

System and method for characterizing a web page using multiple anchor sets of web pages

A method and device are provided for taking a database snapshot using a fuzzy snapshot. In one example, the method includes starting and maintaining a transaction log of the database, starting and maintaining a fuzzy snapshot of the database without applying any hindering locks to the database, and restoring the database by applying the transaction log to the fuzzy snapshot.

Restoring a database using fuzzy snapshot techniques

The present invention provides a method and system of accessing at least one target file in a computer system with an operating system with file locking implemented with byte-range locking. In an exemplary embodiment, the method and system include obtaining a set of handles that corresponds to a set of all files that are open in the computer system, determining within the kernel of the operating system a set of file identifiers that corresponds to the set of handles, identifying from the set of file identifiers a file identifier that corresponds to the target file, sending the identified file identifier to the kernel, initializing within the kernel file caching for the identified file identifier, and requesting within the kernel the cache manager of the operating system to obtain by using the identified file identifier a region of the target file from the file system driver of the operating system.

Method and system of accessing at least one target file in a computer system with an operating system with file locking implemented with byte-range locking

Growing community of inventors

San Jose, CA, United States of America

Benjamin Clay Reed