Presenter: Dr. Jonathan Koomey , Consulting Professor, Stanford University, and Project Scientist, Lawrence Berkeley National Laboratory

Monday, May 17
9:35-10:05 am, Sutton Complex

View the Presentation

In this keynote, Jonathan Koomey will discuss some of the economic factors making cloud computing an ever-more attractive proposition for delivering computing services. These factors include economies of scale and diversity of users compared to smaller in-house data centers. While not all applications are suitable for the cloud, the economic advantages of these centralized computing installations will make them increasingly attractive for businesses seeking computing services at the lowest total cost.

Presenter: Amy Spellmann, President, Optimal Innovations

Monday, May 17
2:40-3:10 pm, Nassau B

View the Presentation

Cloud computing is maturing, becoming a viable alternative to classic on-premise IT. Cloud facilitates scalability, promising lower fixed and variable costs while supporting enterprise growth. The scalability benefits and cost savings can be achieved through on-demand infrastructure provisioning and reduced on-premise energy consumption. The benefits are compelling; however, a quantitative analysis is required.

This presentation will describe a methodology for predicting performance, energy and cost for expanding on-premise IT into the cloud.  The findings reported in this presentation are documented in "Leveraging the Cloud for Green IT: Predicting the Energy, Cost and Performance of Cloud Computing" (authors: Amy Spellmann, Optimal Innovations; Richard Gimarc, Hyperformix, Inc.; Mark Preston, RS Performance), recipient of a 2009 Computer Measurement Group (CMG) Best Paper Award.

Presenter: Pitt Turner, Executive Director, Uptime Institute

Monday, May 17
8:30-9:00 am, Sutton Complex

View the Presentation

In the near term, the greatest opportunity for data center managers and executives is to take back control of their jobs! Some have, but most have not. In spite of myriads of opportunities, real pressures to change, solid accomplishments of industry leaders, and volumes of best practices, there remains much to do. This presentation will ground us in the importance of business requirements and lead us forward to implementation of proven upgrades.

Panelists include: John Abbott, The 451 Group; Ken Baker, Hewlett Packard; Dr. Albert Esser, Dell; Dr. Jonathan Koomey, Lawrence Berkeley National Laboratory (Moderator)

Tuesday, May 18
11:15-11:50, Sutton Complex

Changes in IT over the past decade have made many data centers obsolete. This roundtable will discuss how IT innovations in the coming decade will drive changes, and how data center owners can ensure today's data centers are still fit for purpose in 2020.

Presenter:  Dr. Werner Vogels, Chief Technology Officer, Amazon.com

Tuesday, May 18
9:20-9:50 am, Sutton Complex

View the Presentation

Building the right infrastructure that can scale up or down at a moment's notice can be a complicated and expensive task, but it's essential in today's business landscape. This applies to an enterprise trying to cut-costs, a young business unexpectedly saturated with customer demand, or a start-up looking to launch. There are many challenges when building a reliable, flexible architecture that can manage unpredictable behaviors of today's internet business. This presentation will review some of the lessons learned from building one of the world's largest distributed systems; Amazon.com. The focus will be on state management which is one of the dominating factors in the scalability, reliability, performance and cost-effectiveness of the overall system.

Panelists include: William Fellows, The 451 Group; James Houghton, Adaptivity; Dan Kusnetzky, The 451 Group (Moderator); Mike Manos, Nokia

Wednesday, May 19
11:30-12:00 pm, Nassau A

How do virtualization/cloud initiatives tie to the data center facility? How can you foresee what application failures a hardware outage might cause, when applications may "float" between servers on layers logical infrastructure? How do you determine what hardware failures an application might cause, if lots of applications suddenly shift servers and create a hot spot in a single rack?

In this session, experts discuss the physical infrastructure issues related to virtualization and cloud paradigms.

Panelists include: William Fellows, The 451 Group (Moderator); Dr. Jonathan Koomey, Consulting Professor, Stanford University, and Project Scientist, Lawrence Berkeley National Laboratory; Antonio Piraino, Vice President and Research Director, Tier 1 Research; Pitt Turner, Executive Director, Uptime Institute

Monday, May 17
2:05-2:35 pm, Nassau A

Organizations requiring more data center capacity or increased IT service have multiple options. They can build and operate their own facility, co-locate, or contract with providers of hosting, managed services, or cloud services. Each option has its own advantages and disadvantages in terms of energy efficiency, availability, cost, flexibility, performance, and risk.

In this session, independent opinionated experts explain how they evaluate the options, taking into account an organization's business requirements, the technical opportunities, Service Level Agreements, and the limitations of each option. Expect deep insights and a spirited debate.

Presenter:  Rachel Chalmers, Research Director, Infrastructure Management, The 451 Group

Wednesday, May 19
10:10-10:40 am, Morgan

View the Presentation

Competition from third-party cloud, PaaS, and hosting services are forcing in-house IT to behave more like a service provider to its internal customers. Virtualization and "internal cloud" are the enabling technologies. Can traditional in-house IT shops stay competitive and meet the needs of their organizations, or will outsourcing soon become the norm?

Presenter: Dr. Max Henrion, Chief Executive Officer, Lumina Decision Systems

Tuesday, May 18
3:20-3:50 pm, Morgan

View the Presentation

Data center capacity planners must grapple with major uncertainties: How rapidly will demand increase for computing cycles, storage, and bandwidth? Will server performance continue to improve according to Moore's Law? Each of these uncertainties affects our sizing decisions; the number of servers, cooling and power capacity, and even the building footprint. The best way to cope with such uncertainty is to preserve flexibility as far as possible. We will show how flexible, capacity planning strategies can reduce TCO and lifecycle energy use in the face of large uncertainties, using Lumina's Analytica Datacenter Capacity Planning Tool (ADCAPT), on a range of typical data center scenarios.

Presenter: Stephen Oliver, Vice President, Marketing & Sales, V•I Chip, Vicor Corporation

Wednesday, May 19
1:35-2:05 pm, Gibson

View the Presentation

High Performance Computing (HPC) processors and memory loads continue to increase power demands at lower voltages while performance density and total cost of ownership challenges increase. Efficient high voltage (AC or DC) distribution and power conversion to the loads - whether power processors, graphic processing units, x86 systems, memory or solid-state storage - are essential to maximize data center metrics such as MegaFLOPS / Watt or PUE. This presentation highlights Factorized Power Architecture solutions for the latest HPC systems, converting from high voltage to sub-1V loads.

Panelists include: Chris Kurkjian, Hewlett Packard; Bob Morris, Corgan Associates; Dean Nelson, eBay; Neil Rasmussen, APC by Schneider Electric; Vince Renaud, Uptime Institute Professional Services (Moderator)

Monday, May 17
9:35-10:05 am, Nassau Suite

Join some of the brightest minds in the industry to get their perspectives on the future of data center design for high performance, low cost, and business agility. Topics covered may include

• intelligent energy management
• new measures of efficiency and productivity beyond PUE
• tighter IT-facilities integration (including the effects of virtualization on physical infrastructure)
• modular and scalable architectures
• remote management
• designs for rapid deployment
• mixed densities and tier levels at a single site
• standardized (maybe even "cookie cutter") designs.

Presenter: Dean Nelson, Senior Director of Global Data Center Services, eBay

Tuesday, May 18
12:45-1:15 pm, Sutton Complex

View the Presentation

eBay manages approximately 90 million active users in 32 countries speaking 7 languages, 24/7/365. To date, our marketplace has over a billion photos and over 210 million active listings. Every day on average there are 8 million new listings, over 90 billion database calls, 6 billion URL hits, 4 billion committed transactions, and 8 Petabytes of storage processing 50TeraBytes of new incremental data.

Our company had a 98% correlation between site infrastructure cost and business growth, meaning as listings increased we had a linear increase in overall site operating costs. The four-year innovation strategy that coupled technology refresh with overall data center efficiency broke this trend by absorbing a doubling of our overall infrastructure capacity/capability while keeping the operating costs nearly flat year-over-year in the world's largest on-line marketplace.

 

Learn how Panduit's eco-system of industry-leading technology partners including Cisco, IBM, and Emerson Network Power use Panduit's Unified Physical Infrastructure (UPI) deployment that integrates Smart Data Centers, Connected Buildings, and Industrial Automation solutions to create an optimized physical infrastructure that lowers risk and costs, while improving flexibility and sustainability.

In these sessions, Panduit and its partners will demonstrate how the management and monitoring of the entire physical infrastructure can now be accomplished through a centralized, or unified, operations center for greater capital and operational efficiencies. Panduit will also unveil their first single cabinet micro data center, populated half with switches and half with servers, to be used within a factory or manufacturing floor.