HiPC - High Performance Computing Conference

Industry, Research and User Symposium

Over the years HiPC has been bringing together researchers in the area of High Performance Computing from both the academia as well as the industry.

Of late there has been an explicit agenda within HiPC to bring together the users / practitioners of HPC including both commercial and non-commercial users, giving them a platform to share their challenges and successes as well as discuss relevent technology issues with the HPC researchers and the HPC vendor community. This is the theme of the Industry, Research and User Symposium at HiPC 2011.

The list of companies participating at HiPC in various ways has been growing to over the years to cover a broad spectrum such as Infosys, IBM, HP, Dell, Netapp, NVIDIA, AMD, Intel, Yahoo, Google, Netweb, Wipro and many others. We expect the list to continue to expand. We expect leaders from the user community to be present as well.

We are currently in the process of finalizing the sessions of the Industry, Research and User Symposium for this year. The sessions planned for this year are:

1. Weather and Climate Modeling Challenges

Session Leaders:
Stefan Heinzel, Garching Computing Centre (RZG), Max Planck Society, Germany
Vipin Chaudhari, University at Buffalo, USA, and CEO of CRL, Pune

Abstract:

Climate and weather simulations are among the most challenging HPC applications. This symposium will bring together HPC experts and researchers from India, US and Europe, in the area of climate and weather modeling. The joint session shall foster a closer collaboration among researchers from these regions. The overall goal of the symposium is to provide a forum for presenting the state-of-the-art in the applications of HPC to climate and weather modeling, for discussing best practices, and for exchanging experiences.

Speakers: Confirmed speakers include

Patrick Jockel, Research Scientist at the Institute for Physics of the Atmosphere, Deutsches Zentrum fur Luft- und Raumfahrt (DLR), Oberpfaffenhofen, Germany

Web: http://www.pa.op.dlr.de/~PatrickJoeckel

Title: The transition of Climate Models into Earth System Models

Abstract: Climate models are continuously further developed into comprehensive Earth System Models (ESMs) representing more and more detailed process descriptions at increasingly higher model resolution. This transition poses several challenges, which are exacerbated by the concurrently changing HPCenvironment. The steadily increasing complexity of the model systems requires new approaches for the overall model design to keep them manageable and extendable, their long life cycles require the application of software engineering techniques in HPC and rely on standardisations. One of the fundamental issues is the methodology to combine ("couple") efficiently different domain and/or process models into a consistent ESM. Often, historically grown legacy codes, or parts of them, and model components based on a variety of numerical methods and parallelisation strategies need to be connected. Several approaches to tackle this issue exist. In this talk we present our own development, the Modular Earth Submodel System, and discuss different coupling approaches, such as the fine granular modularisation for atmospheric chemistry processes, the indirect and direct coupling of an atmosphere - ocean system, and the on-line nesting of a high resolution limited area atmospheric model into a global model.

Ulf Garternicht, German Climate Research Centre (DKRZ), Hamburg, Germany

Web: http://www.dkrz.de/about/Organisation/mitarbeiter/k202059

Title: Technical Infrastructures for Big Data in Climate Research

Abstract: Only recently big data has reached the level of attention of a broader audience since many of the existing data handling infrastructures are hitting the ceiling. Numerical models in climate research are among the biggest producers of large data sets. The German Climate Computing Centre (DKRZ) is handling big data in climate research for more than 25 years. This talk will provide a view on the technical infrastructure that is capable to produce, analyse and archive data at a rate of 10 PB per year.

Shailesh Nayak, Secretary, Ministry of Earth Sciences

Web: www.ncmrwf.gov.in/shailesh-nayak-brief-biodata.pdf

Title: High-Performance Computing for Weather and Climate Prediction

Abstract: The Earth behaves as a single interlinked and self regulating system. It`s subsystems, viz. atmosphere, hydrosphere, cryosphere, geosphere and biosphere function together and their interactions are significant and complex. The energy and material transport within and across subsystems occur from local to global scale in varying space and time. Improved and reliable forecast of weather and climate requires integration of observations using very high resolution dynamical models with realistic representation of all physical processes and their complex non linear interactions. Such models require high performance computing (HPC) facility to a scale of 1-2 Petaflops. Numerical modeling of earth system is not perfect science. There are always trade-offs to be made in terms of computational efficiency, spatial and temporal resolution and sophistication of the physical parameterization and numerical schemes.

The modeling is being done for three purposes. The first one is scientific inquiry. Controlled experiments, varying one condition or parameter at a time, to determine what physics is important and what assumptions are valid. Also, "what if" kinds of experiments are run to understand the behavior of ocean or atmosphere under varying parameters. It is realized by modelers that an essential factor for progress in atmosphere-ocean-earth system simulations is through the enhanced resolution of numerical models to resolve clouds that requires large computing power. Enhancing the scale of models from 100km to 10km resolution results in 1000-fold increase in computation. The improvement of monsoon forecasting at various spatial and temporal ranges as well as climate system modeling would be categorized under this class.

The second use is to interpolate and interpret sparse data in dynamically consistent way to produce `nowcast` or analysis. In addition, it is essential to carry out studies related to observation simulation experiments (OSE), observation system simulation experiments (OSSE) and targeted observation experiments that can guide the planners on the location and type of observations that are crucial for the numerical models. Accordingly observation network can be better formulated. This is highly compute intensive job. Large number of numerical experiments shall have to be carried out to identify these crucial locations where observation network need to be strengthened.

The third use is for predictive modeling viz. weather, ocean state, cyclone track, etc. Short and medium range weather forecast, ocean state and ensemble prediction are in this category. Since weather is an initial value problem, accuracy of the initial condition is as important as the accuracy of the model. Thus, data assimilation is a crucial component of weather predictions. As conventional data coverage is spatially and temporally limited, satellite data provides much better coverage in both space and time. About 90% of the data that goes into the assimilation of any analysis-forecast system comprise of data from satellite and rest from in situ platforms. Typically 40-50GB of satellite data per day is available which is required to be downloaded within 10-15 minutes of its availability.

Ravi Nanjundaiah, Centre for Atmospheric & Oceanic Sciences, Indian Institute of Science

Web: http://caos.iisc.ernet.in/faculty/ravi.html

Title: Challenges in Simulating and Predicting the Indian Monsoon

Abstract: Simulating and predicting the Indian summer monsoon presents challenges from scieintific and computational standpoints. Incorporation of diverse effects such as the interaction of ocean and atmosphere, the occurrence tallest of mountains are some of the challenges faced by climate modellers. Presence of long-distance teleconnections specifically the El-Nino-Monsoon teleconnection and Indian Ocean - Monsoon teleconnections increases the complexity of the problem.

For forecasts on the scale of a season,uncertainities in initial conditions especially in and over the oceans impacts the quality of forecasts. The resolution of models used also has an impact on the quality of forecasts/simulations.

Calculating the complex interactions in a realistic fashion in a reasonable time-frame is a major computational challenge. Simulations and predictions need incorporation of processes such as radiation, cloud formation and boundary-layer interactions which are computationally intensive. For predictions we also need to assimilate data from satellite and conventional sources - these techniques are also very computationally intensive.

Massively parallel techniques have been used for this purpose. Scalability of numerical discretization methods such as the spectral technique is a major issue. Newer methods such as finite volume techniques with better scalability have been implemented into the present generation models.

Efforts at our centre and elsewhere in both these domains will be discussed.

B. N. Goswami, Director, Indian Institute of Tropical Meteorology

Web: http://www.tropmet.res.in/~bng/index.html

Title: Computational Challenges for Improved Forecasts of Monsoon Weather & Climate: Vision of MoES

Abstract: TBD.

2. Computational Biology - Next Generation Sequencing

Session Leaders:
Srinivas Aluru, Iowa State Univ., USA, IIT Bombay, India

Abstract:

The interplay between computing and life sciences is leading to transformative advances in medicine, agricultural biotechnology, and microbial genomics. The goal of this session is to bring together researchers from academia, research labs, and industry in India who are engaged in research in bioinformatics and computational biology, and its applications. This year's session will be focused on high-throughput sequencing technologies, also known as next-generation sequencing (NGS).

These technologies are capable of reading billions of sequences, creating unprecedented challenges for informatics development. The data intensive nature of these technologies is an ideal fit for high performance computing and storage systems.

Speakers: Confirmed speakers include

Rajesh Gokhale, Institute of Genomics and Integrative Biology
Ramesh Hariharan, Strand Genomics
Abhay Jere, Persistent Systems
Andrew Lynn, Jawaharlal Nehru University
Binay Panda, GANIT Labs
Rajgopal Srinivasan, TCS Innovation Labs

3. Computational Fluid Dynamics in Aerospace

Session Leaders:
Rama Malladi, Intel, India
Sunil Sherlekar, Intel, India
Sharan Kalwani, Intel, USA

Abstract:

Simulation in general and Computational Fluid Dynamics (CFD) in particular, have been part of the design flow in industry for quite some time now. Of late, however, the rapidly dropping cost of large computing power has enabled the usage of increasingly sophisticated models for simulation. This has resulted not only in getting simulation results that are closer to experiment, but also - and more interestingly - getting information that cannot be obtained by experiment.

This session aims to get a user perspective of CFD for attendees of HiPC 2011. We have obtained the consent of some of the best people in CFD to speak at this session: Roddam Narasimha & S.M. Deshpande from Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR) Bangalore, Anutosh Moitra from Tata Computational Research Labs. Pune, Sukumar Chakravarthy from Metacomp Technologies California and N. Balakrishnan from Dept. of Aerospace Engineering, Indian Institute of Science (IISc) Bangalore. Given the growing importance of the Aerospace industry in India, the special focus of this session will be on Aerospace.

Speakers: Confirmed speakers include

Roddam Narasimha, Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR)

Title: Indian Aeronautics at Another Cross Roads?

Abstract: This talk will present a broad-brush personal view of the speaker of the current aeronautical scene in India. During the 1990s, India was the hottest technology centre in Asia as the Light Combat Aircraft was being developed. Since then, globalization has made inroads into Indian aeronautics and an increasing number of foreign companies have set up R & D centres in the country. A major new initiative appears to be in the offing in civil aeronautics. The talk will argue that it is possible that we are in the middle of a major change in the involvement of the private sector which, although even now is only a small player in Indian aeronautics, may grow significantly in the coming years.

Sukumar Chakravarthy, Metacomp Technologies, Inc.

Bio :Sukumar Chakravarthy is the Founder and President of Metacomp Technologies, Inc. He obtained his B. Tech. in Aeronautical Engineering from IIT Madras and M.S. & Ph.D. in Aerospace Engineering from Iowa State University, where he conducted a part of his research at NASAs Ames and Langley Research Centres. He was with Rockwell International Science Centre for 15 years, where he led the CFD team. In 1994, he founded Metacomp Technologies, a small business devoted to research, development, dissemination and support of effective computational methodologies in a variety of disciplines. He is internationally recognized for his contributions to CFD and computational sciences and has several publications to his credit. The simulation software that he helped develop is being used by leading organizations around the world in the analysis and design of aerospace, automotive and hydrodynamic vehicles and devices. He has worked at Stanford University as a Research Associate and has been an Adjunct Faculty at the University of California at Los Angeles for several years. He was awarded Engineer of the Year at Rockwell in 1989 and, in 2008, was recognized as a Distinguished Alumnus by Iowa State University.

Title: Bridging the User-Architecture Divide: Adapting Algorithms to Architectures.

Abstract: Reaching into more than three decades of experience with CFD research (algorithms, models and methods), development (codes), and applications (users and usage), the author will track down their relationship to hardware and software architectures over those decades. Adapting computational algorithms to architectures in a manner that achieves increased effectiveness for the user has been an ever-changing landscape that nevertheless finds certain trends repeating themselves in cycles. The author will provide a concise discussion of relevant facts and observations useful to the CFD community as well as the computer industry and academia.

Roddam Narasimha & Suresh Deshpande, Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR)

Bio: Roddam Narasimha was educated at the University College of Engineering and IISc and obtained his PhD from the California Institute of Technology. From 1962 to 1999 he was at IISc in the Department of Aerospace Engineering and in the Centre for Atmospheric and Oceanic Sciences. He was the Director of the National Aerospace Laboratories from 1984 to1993 and of the National Institute of Advanced Studies from 1997 to 2004. His research has chiefly been in aerospace and atmospheric fluid mechanics, which include laminar-turbulent transition, re-laminarisation of turbulent flows, bursting in turbulent boundary layers, memory in free shear flows and shock waves and convective atmospheric boundary layers and the fluid dynamics of cumulus cloud flows. His research has combined experimental, theoretical and computational approaches. He was elected Fellow of the Royal Society in 1992 and Foreign Associate of the US National Academy of Sciences in 2000. His many other honours include the Bhatnagar Prize, the AIAA Fluid Dynamics Award, the Trieste Science Prize and the Padma Bhushan.
Bio: Suresh Deshpande obtained his B.E (Electrical Engineering) from Nagpur University and M.E. (Aeronautical Engineering) & Ph.D. from IISc. He was on the faculty of IISc from 1969 to 2004 and since then is with JNCASR. He was elected Fellow of the Indian Academy of Sciences in 1974, the Indian National Academy of Engineering in 1995, the Aeronautical Society of India in 1998 and the Maharashtra Academy of Sciences in 2001. He was awarded the Biren Roy Trust Award of the Aeronautical Society of India in 1991, the Aeronautical Society of India award for outstanding contributions to rocket & related technologies in 1995 and the Distinguished Alumnus award of the Visvesvaraya National Institute of Technology. He was honoured as an Eminent Professor of Aerospace Engineering by the Aeronautical Society of India. He received the IISc Alumnus award for excellence in research and the DRDO Academic Excellence award in CFD. He was Chairman of the Aeronautical Society of India from 1998 to 2001.

Title: Insight, Prediction, Design: The Role of HPC in Aerospace and Atmospheric Fluid Dynamics.

Abstract: High performance computing is today an integral part of scientific research as well as mathematical modelling and engineering design. Its purposes vary from gaining insight into fundamental problems that are not yet understood (e.g. turbulent flows, cumulus clouds), improving skills in prediction (e.g. aerodynamic parameters of flying systems, weather prediction) and more accurate and reliable design of engineering products or weather modification schemes. Selected examples of work in some of these areas will be described in the talk and likely future opportunities will be briefly touched upon.

N. Balakrishnan, Indian Institute of Science (IISc)

Bio: N. Balakrishnan is an Associate Professor at IISc and also the Founder & Director of S&I Engineering Solutions Pvt. Ltd. He has a Ph.D. from IISc and a Post-Doc from ENSAM, Paris. He was an Assistant Professor at IIT Kanpur from 1996 to 1998. His interests include Aerodynamics, Computational Fluid Dynamics and Industrial Applications. He has more than 30 archival publications and more than 100 publications in conferences and symposia. He has guided the Thesis for seven PhD, six M.Sc. and more than 40 M.E. students.

Title: The "HiFUN" Challenge to HPC.

Abstract: Historically, Aerospace CFD has been a prime mover in the development of HPC. The talk will introduce a typical CFD process and highlight the development of the parallel CFD solver called HiFUN (HIgh resolution Flow solver on UNstructured meshes). The experience of scaling HiFUN from a modest IISc-PACE128 (128 nodes of P3) to the 133 TF CRL-EKA with 14,000 Xeon cores will be brought out. The issues behind achieving good parallel scalability without compromising on the algorithmic scalability and how HPC can directly impact the aerodynamic design process will also be discussed. A more recent study on the scalability of HiFUN, carried out in collaboration with Intel will be presented. The talk will conclude by summarizing the expectations of an average CFD code developer from the HPC community.

Anutosh Moitra, Tata Computational Research Labs (CRL)

Bio: Anutosh Moitra is the head of Aerospace Operations and a Distinguished Scientist of Tatas CRL where he coordinates CRLs involvement in India`s aerospace development programs and leads various research teams to generate innovative solutions to problems in airplane design and performance analysis. He is responsible for formulating and executing CRL`s strategies for technological collaborations with India`s aerospace partners. Before joining CRL in 2011 he was a member of Technical Fellowship at Boeing in the Enabling Technology and Research organization. His pioneering contributions to CFD were the key to the success of many design projects at Boeing. Before joining Boeing, he served as a Principal Scientist at NASA Langley Research Centre. He has received numerous recognitions and awards at Boeing and NASA for leading teams to generate innovative solutions to problems in airplane design and operation.

Title: Aerospace Industrys CFD needs and Demands on HPC Capability Growth.

Abstract: Since their practically simultaneous advent in the 1980s modern CFD and HPC have enjoyed a strong interrelationship. Due to its need for conducting numerous very large scale and fast simulations routinely, the aerospace industry has emerged over the last couple of decades as a principal driver of both CFD and HPC technologies. On the other hand, technological leaps in the HPC industry have delivered orders of magnitude increases in processing power, enabling new CFD technologies capable of simulating fluid phenomena in ever increasing detail, thereby whetting the aerospace industrys appetite for ever more realistic CFD simulations. This make it possible to reduce the number of wind-tunnel tests needed for designing airplanes and may one day eliminate the need for such tests altogether. This talk will attempt to give an account of the dynamically evolving CFD needs of the aerospace community and their relationship to projected HPC capability growth.

Stan Posey, NVIDIA

Bio: Stan Posey currently manages the NVIDIA strategy of HPC applications and industry development for a variety of disciplines, with special focus on computational mechanics. Prior to joining NVIDIA in 2009, Mr. Posey has contributed for more than 20 years in HPC applications development including vendor roles at Panasas, SGI, and Control Data Corporation, and engineering roles at CD-adapco and US DOE Oak Ridge National Laboratory. Mr. Posey earned a B.Sc and M.Sc. in Mechanical Engineering from the University of Tennessee, Knoxville, TN, USA.

Title: NVIDIA GPU Computing for CFD Applications

Abstract: This presentation will examine the motivation and trends for GPU-acceleration of computational fluid dynamics (CFD) applications with examples relevant to industry-scale HPC practice. Research organizations and industry, including commercial software vendors, have entered collaborative technical projects with NVIDIA in the development of well-known CFD algorithms. A review will be provided of CFD progress for GPU support and the performance observed both with and without GPU acceleration.

4. Cloud Computing Futures

Session Leaders:
Rama Govindaraju, Google, USA
Ramamurthy Badrinath, HP, India

Abstract:

Cloud Computing is starting to completely change the landscape of computing as we knew it. We are witnessing the rapid growth of mobile hand held devices that provide seamless access to all applications and information from anywhere within a fraction of a second. Supporting all these devices and their applications are massive warehouse scale compute farms which we call the Cloud. These new emerging workloads and computing paradigms are best recognized though the emerging applications such as Facebook, twitter, LinkedIn, iTunes, Amazon, Google, among many others. Many others (such as telecoms) are all building their private clouds via partnership with IBM, HP, NetApp, Wipro, DDN, Microsoft, Dell and others.

In this session we will seek to find insights and answers into the following questions:

Is this changing paradigm for real?
What are the implications of such a paradigm shift?
What should academia/industry/users do to adapt to this changing paradigm?
What are they key challenges that need to be overcome to allow this paradigm to provide maximum value to users, and service providers?
What are each of the participants in the panel doing to help address them?
What efficiencies can be extracted when operating at warehouse scale
Provide a vision for the future of Cloud Computing.

Speakers: Confirmed speakers include

Aruna Ramanan, IBM

Bio: Dr. Aruna Ramanan is a technical leader in High Performance Computing Systems development at IBM. In her 17 years at IBM, she has played a significant role in the development of many generations of clustered systems. She received a PhD in Electrical Engineering from University of Colorado at Boulder in 1993.

Title: Cloud Computing Future: Is it a song that will never end?

Abstract: No paradigm in computing has met with as much skepticism as cloud computing. So far, despite concerns and criticisms, the concept has continued to flourish. In this presentation the author will analyze the dynamics of cloud computing.

Thanunathan Rangarajan, Intel

Bio: Thanunathan Rangarajan is a Platform Software Architect in the Data Center Group in Intel India. Over the last ten years, he has contributed significantly to the architecture and development of server system BIOS and various platform-level technologies. His overall industry experience is of fourteen years. He received a Master's Degree in Electronics and Communication from the Delhi College of Engineering, New Delhi.

Title: Achieving Data Center Dynamism Through Correlation of Node Compute, Power and Thermal Characteristics

Abstract: Data Centers today are planned, built and managed in a static model, whereas their compute workloads are highly dynamic. Scheduling workloads in a manner that is cognizant of the power & thermal characteristics of the Data Center can help achieve balance between the compute requirements and the facilities. This talk will provide an insight into the research work conducted by Intel in this area.

Saji Thoppil, Wipro

Bio: Mr. Saji Thoppil is the General Manager for Enterprise Cloud Computing at Wipro. He is the Chief Architect of Wipro's Public Cloud, the most successful Public Cloud based in India. His leads a team specializing in virtualization, High Performance Computing and Big Data solutions. He is a graduate in Electronics Engineering & carries many industry certifications. His recent work, FluidState Datacenter framework is nominated for the Nascomm innovation award.

Title: Bringing HPC to Enterprise with Cloud

Abstract: Mr. Saji will discuss the Key forecasted Trend on Advanced Analytics, Enterprise Video, Context aware computing, Ubiquitous computing etc. and its impact on today's Enterprise IT canvas. The key role the technologies of HPC & Cloud should play to make this possible.

Thiruvalluvan M G, Yahoo

Bio: Thiruvalluvan M G is an architect with the Cloud Platform Group in Yahoo, Bangalore. He works in cloud infrastructure products and deployment solutions. He has worked in different areas of Internet scale systems in search and advertising before cloud technologies. His interests include large scale distributed computing, deployment automation, software engineering and agile methods for software project execution.

Title: Automating Software Deployments on the Cloud

Abstract: With infrastructure to applications available on demand, cloud poses interesting challenges to the process of software deployments. With agile development methods catching on, companies are moving towards "continuous delivery." The requirement for repeatable and reliable software deployment process has become very important. This talk will discuss the challenges and solutions for automation of large scale software deployments on the cloud.

Phil Wainewright, Boston Limited

Bio: Phil Wainewright is a trusted thought leader in cloud computing, as a blogger, analyst and consultant. He maintains a popular blog on software-as-a-service on ZDNet, serves as vice-president of EuroCloud. At Procullux Ventures, he advises top cloud vendors on messaging, positioning and go-to-market strategy. His forthcoming book, Frictionless Enterprise, describes how forward-looking enterprises are harnessing the cloud for business transformation.

Title: Hardware Choices That Can Make or Break Your Cloud

Abstract: Today's fast-moving business environment is driving enterprise adoption of cloud computing to achieve greater agility, scalability and economy in the IT infrastructure. Starting from first principles of successful cloud architectures, this session will outline the practical steps data center operators can take to fulfill those needs by delivering high-performance, reliable cloud computing with flexible provisioning and efficient, cost-effective operations.

Ramesh Rajagopalan, Dell

Bio: Ramesh Rajagopalan is an Engineering Director at Dell India R&D Center. His group is responsible for delivering high performance computing clusters, database cluster solutions, collaboration solutions and next generation virtualization solutions. His overall industry experience of 21 years. His interests include clustering and performance characterization. He has a BE (Computer Science) from Indian Institute of Science, Bangalore, (1990).

Title: Open Standard Based Approach for Building HPCC

Abstract: This talk will dwell into the challenges of compute and storage domains, the key elements of an HPC solution. The topic will provide insights into the adoption of GPGPU and development in networking and how these are further helping end users adopt HPC. The session will provide examples best practices and tools that would assist the HPC community. The session would provide glimpses into critical factors such as Power Efficiency and Management. It will also provide intersection of technologies and trends for HPC, Cloud and Virtualization.

Vishal Bharat, Mellanox

Bio: Vishal Bharat is Country Sales Manager at Mellanox Technologies Ltd.

Title: Paving the way for Exascale Computing through Convergence of InfiniBand & Ethernet

Abstract: TBD

Gaurav Makkar, NetApp

Bio: Gaurav Makkar is a Technical Director at NetApp, currently driving the technical strategy around Big Data storage solutions, specifically in the areas around Hadoop, No-SQL databases, and wide columnar databases. His interests are in metadata management, information retrieval, scalability, high availability, and performance engineering. He graduated from Jawaharlal Nehru Technological University, Hyderabad, India, in 1993 with a Bachelor of Technology degree in Computer Science and Engineering.

Title: TBD

Abstract: TBD

Vik Malyala, Supermicro

Bio: As a Director of Technical Marketing for Supermicro, Vik Malyala is responsible for bringing Enterprise, Data Center and Application optimized server and storage solutions to market. Vik has over 16 years of experience in enterprise computing area, where he is involved heavily in x86 architecture based servers and chipsets. Vik holds a Masters degree in Electrical Engineering from University of Idaho.

Title: Hardware Design Considerations for Next Generation HPC Solutions

Abstract: Today's Computing solutions are udrnder ever increasing demands to run 24x7 at peak performance processing large amounts of data in a cost effective manner. Design of next generation data center and HPC solutions require constant platform level innovation. New deployments can take full advantage of these innovations to minimize TCO. End users and Systems Integrators can find flexibility and the largest selection of next generation x86 solutions from Supermicro to achieve the right balance between design, efficiency, performance, and cost.

Ashish Dalela, Cisco

Bio: Ashish Dalela has about 15 years of industry experience in networking and telecommunications. He currently works at Cisco Systems on datacenter products on LAN / SAN convergence. Ashish is active on cloud standards in ITU-T and IETF, in areas of architecture and protocols for cloud interoperability.

Title: Interoperable Clouds - Challenges and Requirements

Abstract: This talk will describe the importance of cloud interoperability and the use-cases around interoperability. Interoperability in the Internet has been solved through protocols and architecture standards. This talk will argue that cloud interoperability is an Internet problem and discuss requirements of on cloud protocols and architectures.

Satyendra Rana, TATA CRL

Bio: Dr. Satyendra Rana is the Vice President of Marketing and Business Development at CRL and is responsible for shaping and crystalizing CRL's positioning in the HIgh Performance Computing space and the growth and expansion of CRL business worldwide through strategic partnerships, innovative solutions and business models.

Dr. Rana is a highly accomplished and award-winning IT Executive and Corporate Officer with extensive record of success in product management, professional services, global program management, and startup operations and has special expertise in strategic planning, business-IT alignment, business transformation, change management, service-oriented architecture (SOA), and process innovation.

As founding president of TIE Austin,a leading global entrepreneural network, Dr. Rana mentored, served on the boards, and invested in various technology startups. A former faculty member of Computer Science & Engineering at Indian Institute of Technology, New Delhi and later at Wayne State University in US, Dr. Rana led many funded researchprojects,supervised many Ph.D and graduate students, and published over 50 research papers in international journals and conferences.

Dr.Rana earned his Ph.D in Distributed Computing in 1980 from Indian Institute of Technology, New Delhi, India. He also earned a gold medal for securing first rank in the university during his Bachelor's program in Mathematics (Honours).

Title: Big Data in the Cloud: Few Observations

Abstract: Big Data has become a big buzz word in the industry. More and more companies are mushrooming up funding by VC firms who see enormous potential in this sector. Traditional companies are also retooling their offerings to cater to the buzz. The whole premise of this activity is primarily there is value in data that can be unearthed and monetized. This talk is structured around few observations that help understand ongoing activity in this sector from a pragmatic perspective.

HiPC Industry, Research and User Symposium Co-Chairs

Jigar Halani, Wipro, India
Sharan Kalwani, KAUST, Saudi Arabia
Rama Malladi, Intel, India