High Performance Computing

 

September 11-12 

Day 1 | Day 2

Wednesday, September 12

7:30 am Breakfast Presentation (Sponsorship Opportunity Available)

8:15 Breakout Discussion Groups

Grab a cup of coffee and join one of the discussion groups. These are moderated discussions with brainstorming and interactive problem solving, allowing conference participants from diverse areas to exchange ideas, experiences, and develop future collaborations around a focused topic.

Click here to view dicussion groups

HPC and Virtualization 

9:15 Chairperson’s Remarks

Kevin Davies, Ph.D., Editor-in-Chief, Bio-IT World

9:20 Software-Defined Networking: Directly Programming the Network for Innovation in Science and Business

Dan PittDan Pitt, Ph.D., Executive Director, Open Networking Foundation

At long last, Software-Defined Networking (SDN) turns the network into the servant of the applications. Using concepts and tools consistent with the virtualization of computing, now network owners, enterprises, and departments can directly incorporate specific scientific and business objectives in the instructions governing network operation. In this talk, we will explain the fundamental importance of SDN, show how the OpenFlow protocol makes it possible, and suggest ways in which the Open Networking Foundation and the life sciences community can work together to turn promise into practice.

9:50 Toward a Performant Cloud

Paul GrunPaul Grun, Chief Scientist, System Fabric Works, Inc.

To its users, an idealized cloud appears as an infinitely large and infinitely capable compute and storage resource, but for cloud providers the reality is much different. Whether for organizational, regulatory, technical or historical reasons, the compute and storage resources comprising a cloud are often distributed over large geographical distances, and yet the cloud user expects to interact with computers and storage as though those resources are local to his desktop. For the notion of cloud computing to be successful, the user’s experience must be as close to what he would experience in his home environment as possible. This places great demands on the network and storage resources which are the foundational elements of a successful cloud. In this session, we will describe some recent experimental work in advanced WAN technology that holds the promise of dramatic improvements in the basic infrastructure of the cloud, thus contributing to an improved user experience. These experiments, conducted using the DOE’s ESNet Advanced Network Infrastructure testbed, applied RDMA technology over 40Gb/s WAN links, giving the user direct access to remote datasets over long distances and providing increased WAN efficiency to transport large files when a physical data copy is called for.

10:20 Coffee Break in the Exhibit Hall with Poster Viewing


» Plenary Keynote Session 

11:00 Accelerating Life Sciences with Utility Supercomputing and Cloud HPC

Jason Stowe, CEO, Cycle Computing, LLC.

This talk will discuss various use cases for running common life science tasks, including NextGen Sequencing, Genomics, bioinformatics, molecular modeling, GWAS, PK/PD, and clinical trial simulations using Utility Supercomputing and cloud HPC. Real world use cases, from Cycle's Top 10 Pharma and BioTech clients will be covered.

IBM11:30 The Value of Private Cloud for Life Sciences

Jeff Karmiol, Senior Product Manager, HPC Cloud & Analytics, IBM Platform Computing
While cloud computing presents many opportunities for organizations to improve efficiency and realize bottom-line business advantages, there are also barriers to adopting public cloud-based solutions – such as security, licensing and pricing. This presentation will delve into the value and best practices of deploying a private cloud and how to leverage cloud technology to fully utilize resources within your genomic or life science organization, as well as examples of when it makes sense to expand to public cloud services. 

12:00 pm Forget About the Clouds, Shoot For the MOON

Wu-chun Feng photoWu-Chun Feng, Director, SyNeRGy Laboratory; Associate Professor, Department of Computer Science, Department of Electrical & Computer Engineering, Virginia Polytechnic Institute and State University

Project MOON, short for MapReduce On Opportunistic eNvironments, proactively harvests the unused compute cycles of volatile computing resources (such as desktop or laptop computers in a company) and combines them with a small number of dedicated computing resources to provide the illusion of a robust supercomputer.  MOON computing differs from cloud computing in three ways:  (1) the adopted programming model for the MOON environment is MapReduce, the same technology that Google built to power its search engine, (2) MOON computing can be effectively realized on largely unreliable compute resources, and (3) MOON makes use of the idle or unused compute cycles within one's own institution or enterprise, i.e., "in-sourcing" computing, versus "out-sourcing" the computation to cloud providers such as Amazon or RackSpace.  The latter point substantially enhances the return on investment on institutional computing resources, whether they be in the classroom or out in the battlefield.

12:30 Close of High Performance Computing



Day 1 | Day 2



Premier Sponsors

Annai Systems

Aspera 
 

Cycle Computing
 

DNAnexus
 

IBM 

Official Media Partner

Bio-IT World 

View All Sponsors 

View Media Partners 

Cloud Usage Study 

* IBM and the IBM logo are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide.