Thomas Phan - San Jose CA, US Kavitha Ranganathan - New York NY, US Radu Sion - Sound Beach NY, US
International Classification:
G06K 15/00
US Classification:
358 116
Abstract:
The embodiments of the invention provide a method, service, computer program product, etc. of co-scheduling job assignments and data replication in wide-area systems using a genetic method. A method begins by co-scheduling assignment of jobs and replication of data objects based on job ordering within a scheduler queue, job-to-compute node assignments, and object-to-local data store assignments. More specifically, the job ordering is determined according to an order in which the jobs are assigned from the scheduler to the compute nodes. Further, the job-to-compute node assignments are determined according to which of the jobs are assigned to which of the compute nodes; and, the object-to-local data store assignments are determined according to which of the data objects are replicated to which of the local data stores.
Automated And Dynamic Management Of Query Views For Database Workloads
The embodiments of the invention provide a method, program storage device, etc. for automated and dynamic management of query views for database workloads. More specifically, a method begins by executing queries, which includes accessing a set of data tables for each of the queries. During the executing of the queries, the method accesses a required data table from a cache if the required data table is present in the cache and creates the required data table if the required data table is not present in the cache. The accessing of the required data table from the cache has a lower processing cost than accessing the required data table from a base table. Also during the executing of the queries, created data tables are stored in the cache, wherein one or more of the created data tables are removed from the cache when the cache becomes full.
System And Method For Automating And Scheduling Remote Data Transfer And Computation For High Performance Computing
Ramesh Natarajan - Pleasantville NY, US Thomas Phan - San Jose CA, US Satoki Mitsumori - Tokyo, JP
International Classification:
G06F 9/44
US Classification:
718102
Abstract:
The invention pertains to a system and method for a set of middleware components for supporting the execution of computational applications on high-performance computing platform. A specific embodiment of this invention was used to deploy a financial risk application on Blue Gene/L parallel supercomputer. The invention is relevant to any application where the input and output data are stored in external sources, such as SQL databases, where the automatic pre-staging and post-staging of the data between the external data sources and the computational platform is desirable. This middleware provides a number of core features to support these applications including for example, an automated data extraction and staging gateway, a standardized high-level job specification schema, a well-defined web services (SOAP) API for interoperability with other applications, and a secure HTML/JSP web-based interface suitable for non-expert and non-privileged users.
Automated And Dynamic Management Of Query Views For Database Workloads
INTERNATIONAL BUSINESS MACHINES CORPORATION - ARMONK NY
International Classification:
G06F 7/10 G06F 17/30
US Classification:
707 2, 707E17017
Abstract:
The embodiments of the invention provide a method, program storage device, etc. for automated and dynamic management of query views for database workloads. More specifically, a method begins by executing queries, which includes accessing a set of data tables for each of the queries. During the executing of the queries, the method accesses a required data table from a cache if the required data table is present in the cache and creates the required data table if the required data table is not present in the cache. The accessing of the required data table from the cache has a lower processing cost than accessing the required data table from a base table. Also during the executing of the queries, created data tables are stored in the cache, wherein one or more of the created data tables are removed from the cache when the cache becomes full.
Non-Intrusive, Context-Sensitive Integration Of Advertisements Within Network-Delivered Media Content
A system, method and computer program product for integrating an advertisement within network-delivered media content is described. In accordance with one implementation, media content is retrieved for delivery to an end user, wherein the media content type is one of text, graphic, video or audio. The media content is analyzed to identify an advertisement insertion location within the media content. An advertisement is inserted within the media content at the advertisement insertion location, wherein the advertisement type is the same as the media content type. The media content is then delivered to the end user via a network.
A system for selecting an advertisement for display to a user. The system includes a plurality of web properties and an advertisement engine. Each of the web properties include a web interface that may be customized based on user profile data provided by the user. As such, data may be explicitly associated with the user's profile naturally over time as the user interacts with various web pages and sets user preferences. Upon visiting one of the web properties, the web interface may request an advertisement for the advertisement engine to be displayed to the user. The advertisement engine identifies the user and accesses the user profile data for the identified user stored in each of the web properties. The advertisement engine also accesses advertisement target profile data associated with an advertisement and compares the user profile data to the advertisement target profile data to determine whether to display the advertisement to the user.
System And Method For Setting Bid Prices Associated With Digital Advertisements Based On Market Conditions
Systems and methods for setting a bid price associated with a digital ad are disclosed. Generally, an ad campaign management system calculates a segmented budget for an ad campaign comprising a digital ad. The ad campaign management system runs the ad campaign for a defined period of time to determine at least a click-through rate associated with the digital ad. The ad campaign management system examines an actual cost-per-click (“actual CPC”) and a contractual cost-per-click (“contractual CPC”) associated with the digital ad to determine a calculated cost-per-click (“calculated CPC”) associated with the digital ad for a segment of the ad campaign. The ad campaign management system then sets a bid price for the digital ad based on the calculated CPC for a future segment of the ad campaign.
Analyzing Images To Derive Supplemental Web Page Layout Characteristics
Thomas Phan - San Jose CA, US Rajendra Shriwas - Fremont CA, US
Assignee:
YAHOO! INC. - Sunnyvale CA
International Classification:
G06F 7/06
US Classification:
707 3, 707E17108
Abstract:
Methods and apparatus for automatically and dynamically constructing a web page having one or more images as the web page's major component are provided. The image(s) are analyzed based on a first set of rules to determine at least one relevant attribute, such as a relevant color. At least one aesthetic design element for the web page is automatically selected with reference to the at least one relevant attribute based on a second set of rules. The web page is automatically and dynamically constructed by incorporating the at least one aesthetic design element based on the second set of rules. The first set of rules includes at least one rule that governs the analysis of the image(s) and the determination of the relevant attribute(s). The second set of rules includes at least one rule that governs the selection and the incorporation of the aesthetic design element(s) for the web page.
Samsung R&D Center since 2010
Senior Staff Research Engineer
Microsoft 2008 - 2010
Senior Research Software Development Engineer
Yahoo! 2006 - 2008
Technical Yahoo!
IBM Almaden Research Center 2004 - 2006
Postdoc researcher
UCLA Computer Science Department 1997 - 2004
Graduate Student Researcher
Education:
University of California, Los Angeles 1998 - 2002
University of California, Los Angeles 1996 - 1998
University of Washington 2009 - 2014
University of California, Davis
De Anza College 2005 - 2008
Skills:
Distributed Systems Mobile Applications Machine Learning Mapreduce Android C++ R&D Software Engineering Algorithms Signal Processing Natural Language Processing Programming Computational Linguistics Java Data Mining Software Development Genetic Algorithms Research Mobile Devices Sensors Sensor Fusion Mobile Sensing Activity Recognition Hadoop Data Pipelines Location Based Services Accelerometer Maps Gis Wearable Computing Wearables Engineering Management Healthcare Mobile Health Smartphones Data Science Research Management Sentiment Analysis Deep Learning Biomedical Engineering Data Analysis Python Tizen
Certifications:
Hipaa Privacy and Security Hipaa For Business Associate Ii Hipaa For Business Associate Iii Hipaa For Practice Front Office Ii Hipaa For Practice It Professionals Litmos Healthcare Division, September 2016 Litmos Healthcare Division, October 2016