Search

Reproducibility of results is a cornerstone of the scientific method. Scientific computing encounters two challenges when aiming for this goal. Firstly, reproducibility should not depend on details of the runtime environment, such as the compiler version or computing environment, so results are verifiable by third-parties. Secondly, different versions of software code executed in the same runtime environment should produce consistent numerical results for physical quantities. In this manuscript, we test the feasibility of reproducing scientific results obtained using the IllinoisGRMHD code that is part of an open-source community software for simulation in relativistic astrophysics, the Einstein Toolkit. We verify that numerical results of simulating a single isolated neutron star with IllinoisGRMHD can be reproduced, and compare them to results reported by the code authors in 2015. We use two different supercomputers: Expanse at SDSC, and Stampede2 at TACC. By compiling the source code archived along with the paper on both Expanse and Stampede2, we find that IllinoisGRMHD reproduces results published in its announcement paper up to errors comparable to round-off level changes in initial data parameters. We also verify that a current version of IllinoisGRMHD reproduces these results once we account for bug fixes which have occurred since the original publication. [ABSTRACT FROM AUTHOR]

Since simulation became the third pillar of scientific research, several forms of computers have become available to drive computer aided simulations, and nowadays, clusters are the most popular type of computers supporting these tasks. For instance, cluster settings, such as the so-called supercomputers, cluster of workstations (COW), cluster of desktops (COD), and cluster of virtual machines (COV) have been considered in literature to embrace a variety of scientific applications. However, those scientific applications categorized as high-performance computing (HPC) are conceptually restricted to be addressed only by supercomputers. In this aspect, we introduce the notions of cluster overhead and cluster coupling to assess the capacity of non-HPC systems to handle HPC applications. We also compare the cluster overhead with an existing measure of overhead in computing systems, the total parallel overhead, to explain the correctness of our methodology. The evaluation of capacity considers the seven dwarfs of scientific computing, which are well-known, scientific computing building blocks considered in the development of HPC applications. The evaluation of these building blocks provides insights regarding the strengths and weaknesses of non-HPC systems to deal with future HPC applications developed with one or a combination of these algorithmic building blocks. [ABSTRACT FROM AUTHOR]

To deliver the full science potential of the square kilometer array (SKA) telescope, several SKA regional centres (SRCs) will be required to be constructed in different SKA member countries around the world. These SRCs will provide high performance compute and storage, for the generation, of advanced science data products from the basic data streams generated by the SKA science data handling and processing system, critically necessary to the success of the key science projects to be carried out by the SKA user community. They will also provide support to astronomers to enable them to carry out analysis on very large SKA datasets. Construction of such large data centre is a technical challenge for all SKA member nations. In such a situation, each country plans to construct a smaller SRC over the next few years (2022 onwards), known as a proto-SRC. In India, we propose to construct a proto-SRC, which will be used for the analysis of data from SKA pathfinders and precursors with strong Indian involvement, such as uGMRT, Meerkat and MWA. We describe our thinking on some aspects of the the storage, compute and network of the proto-SRC and how it will be used for data analysis as well as for carrying out various simulations related to SKA key science projects led by Indian astronomers. We also present our thoughts on how the proto-SRC plans to evaluate emerging hardware and software technologies and to also begin software development in areas of relevance to SKA data processing and analysis, such as algorithm implementation, pipeline development and data visualisation software. [ABSTRACT FROM AUTHOR]

In this work, we have analyzed the input/output (I/O) activities of Cori, which is a high-performance computing system at the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory. Our analysis results indicate that most users do not adjust storage configurations but rather use the default settings. In addition, owing to the interference from many applications running simultaneously, the performance varies based on the system status. To configure file systems autonomously in complex environments, we developed DCA-IO, a dynamic distributed file system configuration adjustment algorithm that utilizes the system log information to adjust storage configurations automatically. Our scheme aims to improve the application performance and avoid interference from other applications without user intervention. Moreover, DCA-IO uses the existing system logs and does not require code modifications, an additional library, or user intervention. To demonstrate the effectiveness of DCA-IO, we performed experiments using I/O kernels of real applications in both an isolated small-sized Lustre environment and Cori. Our experimental results shows that our scheme can improve the performance of HPC applications by up to 263% with the default Lustre configuration. [ABSTRACT FROM AUTHOR]

High performance computing system, which takes an interconnection network as its infrastructure topology, has been utilized in scientific computing, big data analysis as well as artificial intelligence. With the rapid growth of infrastructure topology (interconnection network) in high performance computing system, the probability of network element malfunction increases dramatically. Generally, the reliability of interconnection network is measured by the traditional (edge) connectivity, which is limited by the minimum degree of the network. In order to overcome the defect, the matroidal connectivity has been newly proposed to extend the edge connectivity. For the regular network G , let B k (G) = { B 1 , B 2 , ... , B k } be a fixed edge partition ordered sequence of G and χ k = (x 1 , x 2 , ... , x k) be a sequence of k integers. For any D ⊆ E (G) , if | D ∩ B i | ≤ x i (i ∈ [ k ]) and G − D is disconnected, then D is named matroidal cut set of G. The matroidal connectivity of G , denoted by λ (G , B k (G)) , is the minimum size of matroidal cut set of G. In this work, we characterize the matroidal connectivity of a class of n -dimensional regular networks G n. To be more specific, we show that λ (G n , B n (G n)) = (k − l + 1) f (l + j) , where f (n) is a function about n , and l , k , n are positive integers (l ≥ 1), j = n − k. As empirical analysis, we directly determine the matroidal connectivity of some well-known interconnection networks, such as hypercube Q n , star graph S T n and alternating group graph A G n. • High performance computing system, which takes interconnection network as its infrastructure topology, has been utilized in scientific computing, big data analysis as well as artificial intelligence. For the regular networks G , we use λ (G , B k (G)) to denote matroidal connectivity of G (where B k (G) = { B 1 , B 2 , ... , B k } is a fixed edge partition ordered sequence of G), which is the minimum size of edge cut set D ⊆ ∪ i = 1 k B i so that G − D is disconnected. In this work, we characterize the matroidal connectivity of a class of n -dimensional regular networks G n. That is, λ (G n , B k (G n)) = (k − l + 1) f (l + j) , where l , k , j (l ≤ k ≤ n) are integers. As empirical analysis, we determine the matroidal connectivity of some kinds of well-known interconnection networks, such as hypercubes (Q n), star graph (S T n) and alternating group graph (A G n). [ABSTRACT FROM AUTHOR]

In this special section we bring you a practice and experience effort in reproducibility for large-scale computational science at SC20. This section includes nine critiques, each by a student team that reproduced results from a paper published at SC19, during the following year’s Student Cluster Competition. The paper is also included in this section and has been expanded upon, now including an analysis of the outcomes of the students’ reproducibility experiments. Lastly, this special section encapsulates a variety of advances in reproducibility in the SC conference series technical program. [ABSTRACT FROM AUTHOR]

The first is the ongoing emergence of multicore and hybrid microprocessor designs, ushering in a new era of computing in which system designers must accept energy usage as a first-order constraint, and application designers must be able to exploit parallelism and data locality to an unprecedented degree. The research areas of cluster, cloud, and data analytics computing, which today provide fundamental infrastructure for all areas of advanced computational science, are being radically transformed by the convergence of at least two unprecedented trends. [Extracted from the article]

Scientific discovery increasingly relies on computation through simulations, analytics, and machine and deep learning. Of these, simulations on high-performance computing (HPC) platforms have been the cornerstone of scientific computing for more than two decades. However, the development of simulation software has, in general, occurred through accretion, with a few exceptions. With an increase in scientific understanding, models have become more complex, rendering an accretion mode untenable to the point where software productivity and sustainability have become active concerns in scientific computing. In this survey paper, we examine a modest set of HPC scientific simulation applications that are already using cutting-edge HPC platforms. Several have been in existence for a decade or more. Our objective in this survey is twofold: first, to understand the landscape of scientific computing on HPC platforms in order to distill the currently scattered knowledge about software practices that have helped both developer and software productivity, and second, to understand the kind of tools and methodologies that need attention for continued productivity. [ABSTRACT FROM AUTHOR]

With exascale-level computation on the horizon, the art of predicting the cost of computations has acquired a renewed focus. This task is especially challenging in the case of iterative methods, for which convergence behaviour often cannot be determined with certainty a priori (unless we are satisfied with potentially outrageous overestimates) and which typically suffer from performance bottlenecks at scale due to synchronization cost. Moreover, the amplification of rounding errors can substantially affect the practical performance, in particular for methods with short recurrences. In this article, we focus on what we consider to be key points which are crucial to understanding the cost of iteratively solving linear algebraic systems. This naturally leads us to questions on the place of numerical analysis in relation to mathematics, computer science and sciences, in general. [ABSTRACT FROM AUTHOR]

• Create a high-level language feature to address the directive-based programmability gap of copying only a subset of data to and from the accelerator. • Evaluate the novel language feature across platforms using CoMD, a proxy Molecular Dynamics application and examine its performance portability. • Analyze the experimental results to determine the effectiveness of the proposed high-level feature on the performance gains as opposed to using a low-level programming framework. As scientific frameworks become sophisticated, so do their data structures. A data structure typically includes pointers and arrays to other structures in order to preserve application's state. In order to ensure data consistency from a scientific application on a modern high performance computing (HPC) architecture, the management of such pointers on the host and the device, has become complicated in terms of memory allocations because they occupy separate memory spaces. It becomes so severe that one must go through a chain of pointers to extract the effective address. In this paper, we propose to reduce the need of excessive data transfer by introducing the idea of pointerchain , a directive that replaces the pointer chains with their corresponding effective address inside the parallel region of a code. Based on our analysis, pointerchain leads to a 39% and 38% reduction in the amount of generated codes and the total executed instructions, respectively. With pointerchain , we have parallelized CoMD, a Molecular Dynamics (MD) proxy application on heterogeneous HPC architectures while maintaining a single portable codebase. This portable codebase utilizes OpenACC, an emerging directive-based programming model, to address the need of memory allocations from three computational kernels in CoMD. Two of the three embarrassingly parallel kernels highly benefit from OpenACC and perform better than the hand-written CUDA counterparts. The third kernel performed 61% of peak performance of its CUDA counterpart. The three kernels are common modules in any MD simulations. Our findings provides useful insights into parallelizing legacy MD software across heterogeneous platforms. [ABSTRACT FROM AUTHOR]

In this paper we present a novel approach to program optimisation based on compiler-based type-driven program transformations and a fast and accurate cost/performance model for the target architecture. We target streaming programs for the problem domain of scientific computing, such as numerical weather prediction. We present our theoretical framework for type-driven program transformation, our target high-level language and intermediate representation languages and the cost model and demonstrate the effectiveness of our approach by comparison with a commercial toolchain. [ABSTRACT FROM AUTHOR]

Over the past four years, the Big Data and Exascale Computing (BDEC) project organized a series of five international workshops that aimed to explore the ways in which the new forms of data-centric discovery introduced by the ongoing revolution in high-end data analysis (HDA) might be integrated with the established, simulation-centric paradigm of the high-performance computing (HPC) community. Based on those meetings, we argue that the rapid proliferation of digital data generators, the unprecedented growth in the volume and diversity of the data they generate, and the intense evolution of the methods for analyzing and using that data are radically reshaping the landscape of scientific computing. The most critical problems involve the logistics of wide-area, multistage workflows that will move back and forth across the computing continuum, between the multitude of distributed sensors, instruments and other devices at the networks edge, and the centralized resources of commercial clouds and HPC centers. We suggest that the prospects for the future integration of technological infrastructures and research ecosystems need to be considered at three different levels. First, we discuss the convergence of research applications and workflows that establish a research paradigm that combines both HPC and HDA, where ongoing progress is already motivating efforts at the other two levels. Second, we offer an account of some of the problems involved with creating a converged infrastructure for peripheral environments, that is, a shared infrastructure that can be deployed throughout the network in a scalable manner to meet the highly diverse requirements for processing, communication, and buffering/storage of massive data workflows of many different scientific domains. Third, we focus on some opportunities for software ecosystem convergence in big, logically centralized facilities that execute large-scale simulations and models and/or perform large-scale data analytics. We close by offering some conclusions and recommendations for future investment and policy review. [ABSTRACT FROM AUTHOR]

Power and energy efficiency are now critical concerns in extreme-scale high-performance scientific computing. Many extreme-scale computing systems today (for example: Top500) have tight integration of multicore CPU processors and accelerators (mix of Graphical Processing Units, Intel Xeon Phis, or Field Programmable Gate Arrays) empowering them to provide not just unprecedented computational power but also to address these concerns. However, such integration renders these systems highly heterogeneous and hierarchical, thereby necessitating design of novel performance, power, and energy models to accurately capture these inherent characteristics. There are now several extensive research efforts focusing exclusively on power and energy efficiency models and techniques for the processors composing these extreme-scale computing systems. This article synthesizes these research efforts with absolute concentration on predictive power and energy models and prime emphasis on node architecture. Through this survey, we also intend to highlight the shortcomings of these models to correctly and comprehensively predict the power and energy consumptions by taking into account the hierarchical and heterogeneous nature of these tightly integrated high-performance computing systems. [ABSTRACT FROM AUTHOR]

Neuroscience initiatives aim to develop new technologies and tools to measure and manipulate neuronal circuits. To deal with the massive amounts of data generated by these tools, the authors envision the co-location of open data repositories in standardized formats together with high-performance computing hardware utilizing open source optimized analysis codes. [ABSTRACT FROM AUTHOR]

Floating-point computation is ubiquitous in high-performance scientific computing, but rounding error can compromise the results of extended calculations, especially at large scales. In this paper, we present new techniques that use binary instrumentation and modification to do fine-grained floating-point precision analysis, simulating any level of precision less than or equal to the precision of the original program. These techniques have an average of 40–70% lower overhead and provide more fine-grained insights into a program’s sensitivity than previous mixed-precision analyses. We also present a novel histogram-based visualization of a program’s floating-point precision sensitivity, as well as an incremental search technique that allows developers to incrementally trade off analysis time for detail, including the ability to restart analyses from where they left off. We present results from several case studies and experiments that show the efficacy of these techniques. Using our tool and its novel visualization, application developers can more quickly determine for specific data sets whether their application could be run using fewer double precision variables, saving both time and memory space. [ABSTRACT FROM AUTHOR]

Power use in data centers and high-performance computing (HPC) facilities has grown in tandem with increases in the size and number of these facilities. Substantial innovation is needed to enable meaningful reduction in energy footprints in leadership-class HPC systems. In this paper, we focus on characterizing and investigating application-level power usage. We demonstrate potential methods for predicting power usage based on a priori and in situ characteristics. Finally, we highlight a potential use case of this method through a simulated power-aware scheduler using historical jobs from a real scientific HPC system. [ABSTRACT FROM AUTHOR]

For more than two decades, the TOP500 list has enjoyed incredible success as a metric for supercomputing performance and as a source of data for identifying technological trends. The project's editors reflect on its usefulness and limitations for guiding large-scale scientific computing into the exascale era. [ABSTRACT FROM AUTHOR]

Creating the next-generation high-performance simulation and analysis environment will be a significant challenge because of power and storage technology trends. Responding to these challenges will require rethinking and reframing how we approach visualization and analysis. A key difference is the need to keep track of a cost per insight in terms of power and storage used. To reduce power and storage costs, an emerging community consensus is that significantly more visualization and analysis should occur in situ--that is, during the simulation run while the data is resident in memory. Using this approach, we need to consider what scientific insights are sought, balanced by power and storage constraints, and then output only the minimal analysis data needed during the simulation run. Emerging research challenges include exploring what types of analysis questions can be answered during postprocessing with compact data products that are generated in situ and what mathematical or statistical techniques will best support this process. [ABSTRACT FROM PUBLISHER]

Computational science and engineering application programs are typically large, complex, and dynamic, and are often constrained by distribution limitations. As a means of making tractable rapid explorations of scientific and engineering application programs in the context of new, emerging, and future computing architectures, a suite of “miniapps” has been created to serve as proxies for full scale applications. Each miniapp is designed to represent a key performance characteristic that does or is expected to significantly impact the runtime performance of an application program. In this paper we introduce a methodology for assessing the ability of these miniapps to effectively represent these performance issues. We applied this methodology to three miniapps, examining the linkage between them and an application they are intended to represent. Herein we evaluate the fidelity of that linkage. This work represents the initial steps required to begin to answer the question, “Under what conditions does a miniapp represent a key performance characteristic in a full app?” [ABSTRACT FROM AUTHOR]

Wormhole routing is a popular routing technique used in network-on-chip. It is efficient but susceptible to deadlock, while deadlock will significantly degrade the network performance of NoC. Most existing adaptive wormhole routings avoid deadlock by reducing the degree of adaptiveness and thus sacrificing network performance. In this paper, we address both deadlock and network performance issues jointly, and propose a probabilistic odd-even (POE) routing algorithm that achieves the minimum packet delivery delay. The proposed POE dynamically adjusts the probabilities of constrained turns that may lead to deadlocks according to the current network conditions, and uses an efficient deadlock detection and recovery scheme when a deadlock happens. By adopting constrained turns adaptively to the network status, it not only reduces the frequency of deadlock and allows the network to be swiftly recovered when it occurs, but also greatly improves the degree of adaptiveness to obtain high network performance. Experimental results show that our method achieves a significant performance improvement both in terms of network throughput and average packet latency compared with the existing methods such as XY, odd-even, abacus turn model and fully adaptive routing algorithm while it only has moderate energy consumption. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) interferometry has rapidly evolved in the last decade and can be considered today as a mature technology, which incorporates computationally intensive and data-intensive tasks. In this paper, a perspective on the state-of-the-art of high performance computing (HPC) methodologies applied to spaceborne SAR interferometry (InSAR) is presented, and the different parallel algorithms for interferometric processing of SAR data are critically discussed at different levels. Emphasis is placed on the key processing steps, which typically occur in the interferometric techniques, categorized according to their computational relevance. Existing implementations of the different InSAR stages using diverse parallel strategies and architectures are examined and their performance discussed. Furthermore, some InSAR computational schemes selected in the literature are analyzed at the level of the entire processing chain, thus emphasizing their potentialities and limitations. Therefore, the survey focuses on the inherent computational approaches enabling large-scale interferometric SAR processing, thus offering insight into some open issues, and outlining future trends in the field. [ABSTRACT FROM AUTHOR]

Abstract: Molecular dynamics modeling has provided a powerful tool for simulating and understanding diverse systems – ranging from materials processes to biophysical phenomena. Parallel formulations of these methods have been shown to be among the most scalable scientific computing applications. Many instances of this class of methods rely on a static bond structure for molecules, rendering them infeasible for reactive systems. Recent work on reactive force fields has resulted in the development of ReaxFF, a novel bond order potential that bridges quantum-scale and classical MD approaches by explicitly modeling bond activity (reactions) and charge equilibration. These aspects of ReaxFF pose significant challenges from a computational standpoint, both in sequential and parallel contexts. Evolving bond structure requires efficient dynamic data structures. Minimizing electrostatic energy through charge equilibration requires the solution of a large sparse linear system with a shielded electrostatic kernel at each sub-femtosecond long time-step. In this context, reaching spatio-temporal scales of tens of nanometers and nanoseconds, where phenomena of interest can be observed, poses significant challenges. In this paper, we present the design and implementation details of the Purdue Reactive Molecular Dynamics code, PuReMD. PuReMD has been demonstrated to be highly efficient (in terms of processor performance) and scalable. It extends current spatio-temporal simulation capability for reactive atomistic systems by over an order of magnitude. It incorporates efficient dynamic data structures, algorithmic optimizations, and effective solvers to deliver low per-time-step simulation time, with a small memory footprint. PuReMD is comprehensively validated for performance and accuracy on up to 3375 cores on a commodity cluster (Hera at LLNL-OCF). Potential performance bottlenecks to scalability beyond our experiments have also been analyzed. PuReMD is available over the public domain and has been used to model diverse systems, ranging from strain relaxation in Si–Ge nanobars, water–silica surface interaction, and oxidative stress in lipid bilayers (bio-membranes). [Copyright &y& Elsevier]

The Baker Hughes high-performance computing environment is deploying a three-tier approach to providing technical computing resources as an alternative to typical high-performance computing. Baker Hughes uses a complete spectrum of computer resources that break from the standards of high-density, dedicated resources. This environment is intended to provide a more flexible and efficient approach that is responsive to business needs. The design has been proven in production systems in the United States and the Russian Federation since the second quarter of 22001100 with multiple proprietary and in-house scientific applications. This paper will discuss a computing infrastructure that Baker Hughes has built to better adapt to meet changing requirements in high-performance computing. [ABSTRACT FROM AUTHOR]

The article reports on the Sunway TaihuLight and Tianhe-2 supercomputers from China whose performance and supercomputing capacity is said to have overtaken the lead rank of the U.S. in 2016. Data presented compares the supercomputing capacity rankings of China and the U.S. from 2000 to 2016. It also discusses efforts of the Chinese government to support advances in life sciences and manufacturing design via supercomputing.

The article focuses on the need to have a global mechanism for planning and coordinating technologies necessary for working on the high performance computing (HPC) systems, across the open-source community. International Exascale Software Project (IESP) is stated to work across the community to identify critical lacunae in the overall software environment and spot opportunities for beneficial integration into the HPC. Also discussed in the article are the benefits of the project.

High-performance simulations in computational science often involve the combined software contributions of multidisciplinary teams of scientists, engineers, mathematicians, and computer scientists. One goal of component-based software engineering in large-scale scientific simulations is to help manage such complexity by enabling better interoperability among codes developed by different groups. This paper discusses recent work on building component interfaces and implementations in parallel numerical toolkits for mesh manipulations, discretization, linear algebra, and optimization. We consider several motivating applications involving partial differential equations and unconstrained minimization to demonstrate this approach and evaluate performance. [Copyright &y& Elsevier]

In this paper, we introduce SciANN, a Python package for scientific computing and physics-informed deep learning using artificial neural networks. SciANN uses the widely used deep-learning packages TensorFlow and Keras to build deep neural networks and optimization models, thus inheriting many of Keras's functionalities, such as batch optimization and model reuse for transfer learning. SciANN is designed to abstract neural network construction for scientific computations and solution and discovery of partial differential equations (PDE) using the physics-informed neural networks (PINN) architecture, therefore providing the flexibility to set up complex functional forms. We illustrate, in a series of examples, how the framework can be used for curve fitting on discrete data, and for solution and discovery of PDEs in strong and weak forms. We summarize the features currently available in SciANN, and also outline ongoing and future developments. • We introduce a new Python package, SciANN, for scientific computing and physics-informed deep learning. • SciANN is built on top of Tensorflow and Keras, therefore inheriting their high-performance computing capabilities. • We show how to use SciANN for curve-fitting, solution of PDEs in strong and weak form, and model inversion. [ABSTRACT FROM AUTHOR]

• We introduce a hierarchical task parallel programming model for distributed memory systems. • We show that the new model provides both flexibility and performance. • We use the model to implement a Cholesky factorization and a solver for the shallow water equations. • We have compared our implementation with other frameworks and shown that it is competitive. Current high-performance computer systems used for scientific computing typically combine shared memory computational nodes in a distributed memory environment. Extracting high performance from these complex systems requires tailored approaches. Task-based parallel programming has been successful both in simplifying the programming and in exploiting the available hardware parallelism for shared memory systems. In this paper we focus on how to extend task-parallel programming to distributed memory systems. We use a hierarchical decomposition of tasks and data in order to accommodate the different levels of hardware. We test the proposed programming model on two different applications, a Cholesky factorization, and a solver for the Shallow Water Equations. We also compare the performance of our implementation with that of other frameworks for distributed task-parallel programming, and show that it is competitive. [ABSTRACT FROM AUTHOR]

We advocate domain‐specific virtual processors (DSVP) as a portability layer for expressing and executing domain‐specific computational workloads on modern heterogeneous HPC architectures, with applications in quantum chemistry. Specifically, in this article we extend, generalize and better formalize the concept of a domain‐specific virtual processor as applied to scientific high‐performance computing. In particular, we introduce a system‐wide recursive (hierarchical) hardware encapsulation mechanism into the DSVP architecture and specify a concrete microarchitectural design of an abstract DSVP from which specialized DSVP implementations can be derived for specific scientific domains. Subsequently, we demonstrate, an example of a domain‐specific virtual processor specialized to numerical tensor algebra workloads, which is implemented in the ExaTENSOR library developed by the author with a primary focus on the quantum many‐body computational workloads on large‐scale GPU‐accelerated HPC platforms. [ABSTRACT FROM AUTHOR]

An introduction is presented in which the editor discusses various reports within the issue on topics including the workshop on clusters, cloud and data for scientific computing (CCDSC), scientific parallel computing, and high performance computing.

This article introduces the special issue on Hot Chips. [ABSTRACT FROM PUBLISHER]