International Business Machines Corporation

Google Inc.

International Business Machines Corporatio

Sridhar Rajagopalan

Eugene Jon Shekita

Andrew Stephen Tomkins

Shanmugasundaram Ravikumar

Roger C Raphael

Kevin Scott Beyer

Marcus Felipe Fontoura

Prabhakar Raghavan

Jason Yeong Zien

Harish Deshmukh

Bruce Gilbert Lindsay

Paul Sherwood Taylor

Byron Edward Dom

Andreas Neumann

David Andrew Gibson

Soumen Chakrabarti

Daniel Gruhl

Matthew Denesuk

Kevin Snow McCurley

Chitrang Shah

Ning Li

Reiner Kraft

Ronald Jason Barber

John A McPherson, Jr

Runping Qi

Anant Deep Jhingran

Adriana Zubiri

Dharmendra S Modha

Nimrod Megiddo

Thomas Guthrie Zimmerman

Hal Jervis Rosen

Robert William Lyle

Jorge Campello De Souza

Joy Aloysius Thomas

Beverly Yang

Tony Leung

Ronald Fagin

Jagdish Chand

Tony Kai-chi Leung

Joerg Meyer

Ran Canetti

Ziv Bar-Yossef

Jon Michael Kleinberg

Amit Sahai

Robert John Tasman Morris

Gurmeet Singh Manku

Mukundha Madhavan Gnana Sekaran

Moses Samson Charikar

Krishna Kumar Kesavan

Junghoo Cho

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling access to APIs. One of the methods includes receiving a request from a client for a computer authorization challenge to access an application programming interface; determining a computer authorization challenge with a difficulty of completion that satisfies a target computational cost for the application programming interface; and providing the computer authorization challenge to the client for access to the application programming interface.

Application programming interface access controls

Automatically extracting profile feature attribute data from event data is disclosed, including: receiving a set of event data; receiving a feature associated with a profiling technique; determining that a plurality of events associated with a user in the set of event data corresponds to a first attribute corresponding to the feature, wherein the first attribute corresponds to a first bin having a first defined value; determining that the plurality of events associated with the user in the set of event data corresponds to a second attribute corresponding to the feature, wherein the second attribute corresponds to a second bin having a second defined value; and creating a user record corresponding to the user indicating presence in the first bin and the second bin.

Automatically extracting profile feature attribute data from event data

Systems, methods and computer program products are disclosed for managing availability of a slave components executing in a distributed system. A master server and a heartbeat protocol may be provided. The heartbeat protocol may define state machines for the master server and/or the slave components. The master server and/or the slave components may act in accordance with the state machines to manage availability of the slave components in the distributed system.

High availability management system for stateless components in a distributed master-slave component topology

Techniques are disclosed for synchronizing a primary data system with an auxiliary data system that processes data for the primary data system. In one embodiment, how current the primary data system and the auxiliary data system are is determined. Requests sent from the primary data system that were not processed by the auxiliary data system are determined. The requests are resent to the auxiliary data system for processing.

Synchronizing an auxiliary data system with a primary data system

Provided are techniques for partitioning a physical index into one or more physical partitions; assigning each of the one or more physical partitions to a node in a cluster of nodes; for each received document, assigning an assigned-doc-ID comprising an integer document identifier; and, in response to assigning the assigned-doc-ID to a document, determining a cut-off of assignment of new documents to a current virtual-index-epoch comprising a first set of physical partitions and placing the new documents into a new virtual-index-epoch comprising a second set of physical partitions by inserting each new document to a specific one of the physical partitions in the second set using one or more functions that direct the placement based on one of the assigned-doc-id, a field value derived from a set of fields obtained from the document, and a combination of the assigned-doc-id and the field value.

Index partition maintenance over monotonically addressed document sequences

In a data mining system, data is gathered into a data store using, e.g., a Web crawler. The data is classified into entities. Data miners use rules to process the entities and append respective keys to the entities representing characteristics of the entities as derived from rules embodied in the miners. With these keys, characteristics of entities as defined by disparate expert authors of the data miners are identified for use in responding to complex data requests from customers.

Knowledge-based data mining system

A dynamic Bloom filter comprises a cascaded set of Bloom filters. The system estimates or guesses a cardinality of input items, selects a number of hash functions based on the desired false positive rate, and allocates memory for an initial Bloom filter based on the estimated cardinality and desired false positive rate. The system inserts items into the initial Bloom filter and counts the bits set as they are inserted. If the number of bits set in the current Bloom filter reaches a predetermined target, the system declares the current Bloom filter full. The system recursively generates additional Bloom filters as needed for items remaining after the initial Bloom filter is filled; items are checked to eliminate duplicates. Each of the set of Bloom filters is individually queried to identify a positive or negative in response to a query. When the system is configured such that the false positive rate of each successive Bloom filter is decreased by one half, the system guarantees a false positive rate of at most twice the desired false positive rate.

Generating and using a dynamic bloom filter

Systems, methods and articles of manufacture are disclosed for managing availability of a component executing in a distributed system. The component may have an address space closed to the distributed system. In one embodiment, the component may be initiated. A state of the component may be analyzed to determine the availability of the component. The determined availability may be transmitted to the distributed system. The component may also be restarted responsive to a request from the distributed system to restart the component.

Managing availability of a component having a closed address space

A cache-aware Bloom filter system segments a bit vector of a cache-aware Bloom filter into fixed-size blocks. The system hashes an item to be inserted into the cache-aware Bloom filter to identify one of the fixed-size blocks as a selected block for receiving the item and hashes the item k times to generate k hashed values for encoding the item for insertion in the in the selected block. The system sets bits within the selected block with addresses corresponding to the k hashed values such that accessing the item in the cache-aware Bloom filter requires accessing only the selected block to check the k hashed values. The size of the fixed-size block corresponds to a cache-line size of an associated computer architecture on which the cache-aware Bloom filter is installed.

System and method for generating a cache-aware bloom filter

Disclosed is an evaluation technique for text search with black-box scoring functions, where it is unnecessary for the evaluation engine to maintain details of the scoring function. Included is a description of a system for dealing with blackbox searching, proofs of correctness, as well experimental evidence showing that the performance of the technique is comparable in efficiency to those techniques used in custom-built engines.

Adaptive evaluation of text search queries with blackbox scoring functions

System and method for generating and using a dynamic bloom filter

Provided is a technique for building an index. A new indexis built and an anchor text tableand a duplicates tableare output using a store, a delta store, and previously generated global analysis computations, wherein the previously generated global analysis computationsinclude an anchor text table, a rank table, and a duplicates table. New global analysis computationsare generated using the anchor text tablei+1, the duplicates table, and the previously generated global analysis computations.

Pipelined architecture for global analysis and index building

Disclosed is a technique for indexing data. For each token in a set of documents, a sort key is generated that includes a document identifier that indicates whether a section of a document associated with the sort key is an anchor text section or a context section, wherein the anchor text section and the context text section have a same document identifier; it is determined whether a data field associated with the token is a fixed width; when the data field is a fixed width, the token is designated as one for which fixed width sort is to be performed; and, when the data field is a variable length, the token is designated as one for which a variable width sort is to be performed. The fixed width sort and the variable width sort are performed. For each document, the sort keys are used to bring together the anchor text section and the context section of that document.

Architecture for an indexer

A system, method, and computer program product to improve XML query processing efficiency with virtual cursors. Structural joins are a fundamental operation in XML query processing, and substantial work exists on index-based algorithms for executing them. Two well-known index features&#x2014;path indices and ancestor information&#x2014;are combined in a novel way to replace at least some of the physical index cursors in a structural join with virtual cursors. The position of a virtual cursor is derived from the path and ancestor information of a physical cursor. Virtual cursors can be easily incorporated into existing structural join algorithms. By eliminating index I/O and the processing cost of handling physical inverted lists, virtual cursors can improve the performance of holistic path queries by an order of magnitude or more.

Virtual cursors for XML joins

Growing community of inventors

Liverpool, United Kingdom

Sridhar Rajagopalan