HPDC'14 will have an achievement award talk and two keynote speakers .
- Krishna V. Palem (Rice University)
Title: Should we break the contract with accuracy at Exascale and beyond? Inexactness for extremely energy-efficient high-performance computing
Abstract: Well over a decade ago, many believed that an engine of growth driving the semiconductor and computing industries, captured nicely by Gordon Moore's remarkable prophecy (Moore's law), was speeding towards a dangerous cliff-edge. Ranging from expression of concern to doomsday scenarios, the exact time when serious hurdles would beset us varied quite a bit-some of the more optimistic warnings giving Moore's law till 2020! Needless to say, this implied major challenges to high-performance computing systems especially as we approach the much desired exascale levels and beyond. When faced with this, I decided to consider a different approach back in 2002-one which suggested falling off the metaphorical cliff as a design choice, but in a controlled manner. This would result in devices, circuits, and architectures, that switch and produce bits of information that can deviate from the "correct" values-the resulting computations are approximate or inexact relative to the original correct specification. However, by being inexact through deliberately compromising accuracy, the resulting hardware designs can be significantly more efficient in the energy they consume, their speed of execution, and physical area needs. This makes inexactness attractive for a range of applications, especially those using floating point representations. In this talk, I will start with the beginnings of our foray into this domain-one that Technology Review labeled as being heretical in their TR10 citation-and give an overview of a range of concepts that groups around the world have been developing since, that embody inexact computing. In particular, I will outline opportunities for realizing extremely energy-efficient inexact floating point architectures and memory systems for high performance computing. As part of the talk, I will distinguish inexactness from traditional notions such as precision of floating point numbers through the concept of information efficiency. Opportunities for deriving substantial energy speed and area benefits through inexactness will be illustrated through a case study of its value in the context of computations for climate models and weather prediction.
Bio: Krishna V. Palem is the Ken and Audrey Kennedy Professor at Rice University with appointments in Computer Science, in Electrical and Computer Engineering and in Statistics. In the past, he founded and directed the NTU-Rice Institute on Sustainable and Applied Infodynamics (ISAID), and is a scholar in the Baker Institute for Public Policy. He was a Moore Distinguished Faculty Fellow at Caltech in 2006-2007, and a Schonbrunn Fellow at the Hebrew University of Jerusalem in 1999, where he was recognized for excellence in teaching. His advisee Suren Talla was awarded the Janet Fabri Prize for outstanding dissertation in 2001, and his related work on the foundations of architecture assembly for designing reconfigurable embedded SoC architectures, developed at Proceler Inc. which he co-founded as a CTO, was a nominee on 2002 for the Analysts choice awards as one of the four outstanding technologies.
A decade ago, he pioneered a novel technology entitled Probabilistic CMOS (PCMOS) for enabling ultra low-energy computing, with his students. In this area, his student Lakshmi Chakrapani's research was recognized through an outstanding dissertation award by Sigma Xi in 2008. PCMOS has also been recognized by a best-paper award at the IEEE-ACM CASES 2006 conference in Seoul, as one of the ten technologies 'likely to change the way we live' by MIT's Technology Review in 2008, and as one of the seven 'emerging world changing technologies' by IEEE as part of its 125th anniversary celebrations in 2009 . Broadly, he has led efforts internationally in the area of embedded systems and their compiler optimizations, for which he has been named a Fellow of the IEEE, the ACM and AAAS. In 2012, Forbes (India) ranked him second on the list of eighteen scientists who are "some of the finest minds of Indian origin". He is the recipient of the 2008 W. Wallace McDowell Award, IEEE Computer Society's highest technical award and one of computing's most prestigious individual honors.
- Karsten Schwan (Georgia Tech)
Title: Data-intensive High End Applications: Challenging Future Systems and Platforms
Abstract: This talk will review the challenges imposed on individual nodes and on the system by data-intensive high end applications. Starting with absorbing the massive I/O volumes of high end simulations, continuing with the need to perform online analytics for such output data, and finding value from 'fast data' processing, results in technical challenges that hit upon memory limitations on individual nodes, seek to leverage new memory technologies like non-volatile memory, attempt to replace I/O stack usage with direct memory accesses, use computation to reduce data movement, and bring forward innovative I/O interfaces enabling such functionality with minimal effort by end users in rewriting their applications. By reviewing the current state of the art and presenting results from both the HPC and enterprise domains, this talk hopes to instigate new research on parallel and distributed solutions for data-intensive applications.
Bio: Karsten Schwan is a Regents' Professor in the College of Computing at the Georgia Institute of Technology. He also a Director of the Center for Experimental Research in Computer Systems (CERCS), with co-directors from both GT's College of Computing and School of Electrical and Computer Engineering. His M.S. and Ph.D. degrees are from Carnegie-Mellon University in Pittsburgh, Pennsylvania, where he conducted research in high performance computing, addressing operating and programming systems support for the Cm* multiprocessor, followed by subsequent research in real-time and distributed systems. His current work ranges from topics in operating systems, to middleware, to parallel and distributed systems, in the enterprise domain and in high performance computing. Specific technical topics include (1) system technologies for future multi-core and multi-machine platforms, (2) efficient methods for managing applications and services in datacenter and cloud computing systems, including new techniques for runtime performance and behavior monitoring and understanding, (3) middleware for high performance data movement, addressing I/O in future petascale machines and QoS-sensitive data streaming in pervasive and wide area systems, and (4) experimentation with representative applications in the HPC, enterprise, and pervasive domains. He can be reached at firstname.lastname@example.org and at www.cc.gatech.edu/~schwan
Rich Wolski (UCSB and Eucalyptus) Achievement award talk
Abstract: Research published over the history of the the Symposium on High-Performance Parallel and Distributed Computing (HPDC) has a discernible impact on current research and technology achievements. Ideas originally explored in HPDC publications, presentations, and discussions now act as the underpinning for large-scale computing systems. At the same time, important HPDC topics remain open questions today for the scalable computing research community.
This talk will discuss some of the influence that HPDC and HPDC-style research has had on the current technology landscape. It will also point to research questions that remain open and discuss possible new research directions for the HPDC community to lead.
Bio: Dr. Rich Wolski is a Professor of Computer Science at the University of California, Santa Barbara (UCSB) and co-founder of Eucalyptus Systems Inc. Having received his M.S. and Ph.D. degrees from the University of California at Davis (while a research scientist at Lawrence Livermore National Laboratory) he has also held positions at the University of California, San Diego, and the University of Tennessee, the the San Diego Supercomputer Center and Lawrence Berkeley National Laboratory. Rich has led several national scale research efforts in the area of distributed systems and is the progenitor of the Eucalyptus open source cloud project.