| |
Seminars in Computer Science |
|
|
| |
Title: CS 201: EDA Algorithm Acceleration on FPGAs and GPUs
Speaker:
SUNIL P. KHATRI, Texas A & M
Date: Tuesday, April 28, 2009
Location: 3400 Boelter Hall
Calendar: Seminars
Contact: Nancy Neymark
|
Title: CS 201: Jaql: Pipes for Analytics in the Cloud
Speaker:
VUK ERCEGOVAC, IBM Research
Date: Thursday, May 28, 2009
Location: 3400 Boelter Hall
Calendar: Seminars
Contact: Nancy Neymark
|
Title: CS 201: Green Computing
Speaker:
Michael Maximilien, IBM Almaden Research Center
Date: Tuesday, June 2, 2009
Time: 4:15 p.m. - 5:45 p.m.
Calendar: Seminars
Contact: Nancy Neymark |
Title: CS 201: Data Cloud: Integrating Structured Information into Web Search
Speaker:
YURY LIFSHITS, Yahoo Research
Date: Thursday, June 4, 2009
Time: 4:15 p.m. - 5:45 p.m.
Calendar: Seminars
Contact: Nancy Neymark |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
Seminar Abstracts |
| |
EDA Algorithm Acceleration on FPGAs and GPUs
Sunil P. Khatri
Texas A & M
Abstract:
In this talk, I will talk about the work being conducted in my group on solving EDA algorithms on FPGAs and GPUs. I will focus on Boolean Satisfiabililty (SAT), and start with a discussion about our FPGA based SAT acceleration engine. The approach is based on partitioning the SAT problem into bins which can be solved on the FPGA. The FPGA implements BCP, implication generation and conflict clause generation in parallel. Inter and intra-bin non-chronological backtracks are done using a GRASP-like engine implemented on the FPGA. I will next discuss our initial efforts on implementing SAT on a GPU. Our approach is a combination of a complete procedure and a fast, incomplete SAT solving approach, retaining the best features of both approaches. I will present the approach and initial results. We have also accelerated other algorithms on the GPU, including fault simulation, Monte Carlo based SSTA, SPICE and fault table generation. I will talk briefly about these as well. I will conclude with a quick summary of the other research efforts that my group is working on. |
Jaql: Pipes for Analytics in the Cloud
Vuk Ercegovac
IBM Almaden Research Center
Abstract:
We introduce Jaql, a query language for the JSON data model. JSON (JavaScript Object Notation) is a popular data format for many Web-based applications because of its simplicity and modeling flexibility. JSON easily models a wide spectrum of data, ranging from homogenous flat data to heterogeneous nested data, and does this in a language-independent format that easily integrates with existing programming languages. We believe that these characteristics make JSON an ideal data format for many Hadoop applications and databases in general. This talk will describe the key features of Jaql and show how it can be used to process JSON data in parallel using Hadoop's map/reduce framework. In addition, we will present several use-cases
from the enterprise setting and discuss our research efforts focused on large scale data analytics. |
Green Computing
Dr. E. Michael Maximilien
IBM Almaden Research Center
Abstract:
IBM has a global campaign focused on making the entire planet smarter and greener. The company strives to be a leader in this area. In this talk, I will discuss the potential impact of cloud computing in making data operations more energy efficient and other ways of making the computing infrastructure greener.
Economies of Scale -- In addition to large quantities of compute resources, cloud providers are able to use economies of scale to provide the compute resources at very low cost. As an example, typical compute clusters for a startup on EC2 cost about $3-$10 per day. Other benefits of cloud computing is that data centers and compute centers are typically energy efficient; this is the case since these centers can be relocated in regions where energy is abundant, cheaper, and greener.
Cloud computing platforms continue the realization of service-oriented computing by providing computing infrastructure-as-a-service. For instance, Amazon's Elastic Compute Cloud (EC2) and Simple Storage Service (S3) allow almost instant access to large quantities of computing resources (CPU, memory, storage, and bandwidth) within seconds of a user's request. Google's App Engine and IBM HiPODS clouds are other examples. |
Data Cloud: Integrating Structured Information into Web Search
Yury Lifshits
Yahoo! Research
Abstract:
In this talk we address two questions: (1) How to use structured data in web search? (2) How to gather structured data? For the first question we identify valuable classes of data, present query classes that can benefit from structured data and describe architecture that combines keyword search with structured search. For the second question we present Data Cloud: An ecosystem of data publishers, search engine (data cloud) and data consumers. We show connection from Data Cloud strategy to classic notion in economics: network effect in two-sided markets. The talk is concluded with descriptions of potential seed projects and a list of research challenges. |
|
| |
|
|
