Get Started for FREE
Sign up with Facebook Sign up with X
I don't have a Facebook or a X account
 Your new post is loading...
Your new post is loading...
Scoop.it!
Gearing up to go full stack, firm spends time on standards, open-source communities... 
No comment yet.
Sign up to comment
Scoop.it!
Browse the GTC conference catalog of sessions, talks, workshops, and more.
Phillip Trotter's insight:
The good folks at NVIDIA have made the recent 324 GTC conference sessions freely available. With topics like Omniverse, Bots, Digital Twins, Autonomous Vehicles and extensive applications of AI , visualization and simulation from experts around the world. They are worth watching and investigating. Enjoy (and thanks NVIDIA!!)
Scoop.it!
The NVIDIA® CUDA® Toolkit provides a comprehensive development environment for C and C++ developers building GPU-accelerated applications. The CUDA Toolkit includes a compiler for NVIDIA GPUs, math libraries, and tools for debugging and optimizing the performance of your applications.
Phillip Trotter's insight:
New CUDA 6 toolkit release with 64-bit arm support, improved CUDA Fortran for scientific aplications, replay features in visual profiler and nvprof and a new BLAS GPU library cublasXT that scales across GPUs. A lot of goodness for compute developers - available here: https://developer.nvidia.com/cuda-downloads
Scoop.it!
An indispensable part of our personal and working lives, computing has also become essential to industries and governments. Steady improvements in computer hardware have been supported by periodic doubling of transistor densities in integrated circuits over the past fifty years. Such Moore scaling now requires ever-increasing efforts, stimulating research in alternative hardware and stirring controversy. To help evaluate emerging technologies and increase our understanding of integrated-circuit scaling, here I review fundamental limits to computation in the areas of manufacturing, energy, physical space, design and verification effort, and algorithms. To outline what is achievable in principle and in practice, I recapitulate how some limits were circumvented, and compare loose and tight limits. Engineering difficulties encountered by emerging technologies may indicate yet unknown limits.
Limits on fundamental limits to computation Via Complexity Digest
Phillip Trotter's insight:
Discussion of limits is key to creating new ideas - Igor Markov's paper is worth reading for exploring lmitations and engineering implications and to trigger off new discussions and ideas. Worth reading.
Scoop.it!
Phillip Trotter's insight:
Jorge Titinger, CEO of SGI has a good article on applications of HPC to Big Data including: Graphing and mapping: HPC-powered data mapping and graphing will lead to greater accuracy in business forecastingPattern visualizations: HPC-powered tools will emerge that can provide an intuitive view of complex data sets, enabling rapid identification of relationships for simple analysisScaling in-memory databases: HPC will allow enterprise in-memory systems to handle larger data workloads — allowing closer to complete data sets (over partial sets) to benefit from real-time analytics while in motionMeta-data: The importance of meta-data will jump dramatically — we’ll see enterprises realize leveraging meta-data analytics for virtualization and relational mapping can yield enhanced accuracy, new business insights and even reveal security threats
With the advent of on demand compute and cloud based processing it will be interesting to see how the HPC companies continue to differentiate from on demand suppliers such as amazon and microsoft. Jorge's 2014 outlook is interesting and more sane than many big data predictions - certainly worth reading.
Scoop.it!
My preferred job title is 'theorist', but that is often too ambiguous in casual and non-academic conversation, so I often settle for 'computer scientist'. Unfortunately, it seems that the overwhelming majority of people equate ...
Phillip Trotter's insight:
Artem Kaznatcheev, a researcher in theoretical computer science - i.e. the ideas that underpin computing - has a wonderful write up of Stephanie Forrest's Stannislaw Ulam lecture at the SFI on using inspiration from Biology to address challenges in Software industry. The Ulam lecture is available in video - but its a few hours long - through seriously worth watching and covers modern developments in genetic programming and other approaches. If you need an abbrieviated write up of the key ideas underpinning the Professor Forrest's lecture - then Artem's write up is an awesomely succinct. Worth reading (and the lectures linked in his article - are worth watching!)
Scoop.it!
From
vimeo
Presented at the MIT Media Lab on April 4, 2013. A peronal preface: http://worrydream.com/MediaForThinkingTheUnthinkable/note.html For more information about…
Phillip Trotter's insight:
Bret Victor talk on media tools for thinking and how representions map to thinking and building associations which help us understand systems. Worth watching.
Scoop.it!
We begin by distinguishing computationalism from a number of other theses that are sometimes conflated with it. We also distinguish between several important kinds of computation: computation in a generic sense, digital computation, and analog computation. Then, we defend a weak version of computationalism—neural processes are computations in the generic sense. After that, we reject on empirical grounds the common assimilation of neural computation to either analog or digital computation, concluding that neural computation is sui generis. Analog computation requires continuous signals; digital computation requires strings of digits. But current neuroscientific evidence indicates that typical neural signals, such as spike trains, are graded like continuous signals but are constituted by discrete functional elements (spikes); thus, typical neural signals are neither continuous signals nor strings of digits. It follows that neural computation is sui generis. Finally, we highlight three important consequences of a proper understanding of neural computation for the theory of cognition. First, understanding neural computation requires a specially designed mathematical theory (or theories) rather than the mathematical theories of analog or digital computation. Second, several popular views about neural computation turn out to be incorrect. Third, computational theories of cognition that rely on non-neural notions of computation ought to be replaced or reinterpreted in terms of neural computation.
Neural Computation and the Computational Theory of Cognition Gualtiero Piccinini, Sonya Bahar Cognitive Science http://dx.doi.org/10.1111/cogs.12012 Via Complexity Digest
Phillip Trotter's insight:
Re-reading some of John Holland's work on neural network simulation at present while looking into different models of computation and digital physics, so this is a timely paper. Looks to be an interesting read.
Scoop.it!
The advent of Intel's massively parallel coprocessor will make every server a supercomputer. This week, Intel unveiled its new Xeon Phi coprocessor, which puts an astonishing 50 x86 cores onto a single PCI-connected card. The term "coprocessor" should be understood in context. Every one of the Phi's cores can boot Linux and run any x86 software. However, the card itself needs to plug into a system that has an independent CPU, which basically oversees the Phi's operations. Hence, the coprocessor appellation. The first model to be released in Q1 of next year will have 50 cores, and the follow-up coprocessor slated for release in mid-2013 will have 60 cores. Each processor supports four threads, making for 200 threads for the initial Phi. The cores run at 1.05 GHz and sport a 512-KB L2 cache each. They collectively share 8 GB of GDDR5 memory. They will liely compete with GPU based solutions from NVIDIA and AMD - but with a more familiar programmign model. Things just got very interesting in the high performance compute world. Click on the image or the title to learn more.
Scoop.it!
Seth Llyod is a Professor in the Department of Mechanical Engineering at the Massachusetts Institute of Technology (MIT). His talk, "Programming the Universe", is about the computational power of atoms, electrons, and elementary particles. Via Szabolcs Kósa, The Robot Launch Pad
Scoop.it!
From
arxiv
The increase of existing computational capabilities has made simulation emerge as a third discipline of Science, lying midway between experimental and purely theoretical branches [1, 2]. Simulation enables the evaluation of quantities which otherwise would not be accessible, helps to improve experiments and provides new insights on systems which are analysed [3-6]. Knowing the fundamentals of computation can be very useful for scientists, for it can help them to improve the performance of their theoretical models and simulations. This review includes some technical essentials that can be useful to this end, and it is devised as a complement for researchers whose education is focused on scientific issues and not on technological respects. In this document we attempt to discuss the fundamentals of High Performance Computing (HPC) [7] in a way which is easy to understand without much previous background. We sketch the way standard computers and supercomputers work, as well as discuss distributed computing and discuss essential aspects to take into account when running scientific calculations in computers. Via Frédéric Amblard
|
Scoop.it!
x86 still 'very important' we're told as lid lifted on Arm-based kit...
Phillip Trotter's insight:
For teams needing HPC capabilities this is going to get interesting very quickly.
Scoop.it!
Nations compete on speed using very different compute architectures.
Phillip Trotter's insight:
Some great insights on the evolving supercomputer market and approaches as chiplet architectures get adopted and promise to deliver significantly improved speeds and capabilities on the next generation of machines - there are a lot of implications for simulation and scientific software developers. Well worth a read.
Scoop.it!
Say what you will about Oracle co-founder and CEO Larry Ellison, but when the software giant bought Sun Microsystems more than four years ago, for $7.4 bil
Phillip Trotter's insight:
It has often been easy to forget following the acquisition of Sun by Oracle - Oracle have continued investing in hardware. The new M7 chip line with 64 sparc cores and Numa interconnect options not only continues to promise speed increases for existing oracle customers but also shows benefits over off the shelf clusters of highly integrated large core systems using optimised connection architectures. Good in depth article - worth reading.
Scoop.it!
Phillip Trotter's insight:
Good article on the differences between big data processing and HPC simulations. Worth reading to see where the two communities focus, worry about and can learn from one another.
Scoop.it!
This was just a brilliant paper, talking about exactly what I found wrong with (yet) current computational models: http://t.co/pxP6MZMa7T
Phillip Trotter's insight:
I remember reading this when first published and its a great paper. Any computational model of human cognition needs to integrate both chemical and eletrical mechanisms into a integrated whole. Great scoop and awesome paper.
Scoop.it!
From
www
Evolutionary information theory is a constructive approach that studies information in the context of evolutionary processes, which are ubiquitous in nature and society. In this paper, we develop foundations of evolutionary information theory, building several measures of evolutionary information and obtaining their properties. These measures are based on mathematical models of evolutionary computations, machines and automata.
Evolutionary Information Theory Information 2013, 4(2), 124-168; http://dx.doi.org/10.3390/info4020124 Via Complexity Digest
Phillip Trotter's insight:
This looks very promising - one for reading list for holidays.
Scoop.it!
An unprecedented trillion-particle simulation, which utilized more than 120,000 processors and generated approximately 350 terabytes of data, pushed the performace capability of the National Energy Research Scientific Computing Center’s (NERSC’s) Cray XE6 “Hopper” supercomputer to its limits.
Phillip Trotter's insight:
To understand the scale of this simulation the team simulated more than two trillion particles for nearly 23,000 time steps with VPIC, a large-scale plasma physics application. The simulation used approximately 80 percent of Hopper’s computing resources, 90 percent of the available memory on each node, and 50 percent of the Lustre scratch file system. In total, 10 separate trillion-particle datasets, each ranging between 30 to 42 terabytes in size, were written as HDF5 files on the scratch file system at a sustained rate of approximately 27 gigabytes per second. This type of simulation will become increasingly necessary for many of the large scale science challenges underway in many areas of physics, geophysics and lifesciences. Click on the image or title to learn more.
Scoop.it!
Application and development of computational methods and tools for modeling and analyzing complex biological systems.
Phillip Trotter's insight:
AS some one who deeply subscribes to Chris Langton's view that the natural stufy of Computing is to stufy computation as its writ across all of nature, this research at Microsoft is deeply interesting (and echoes compaies liek Autodesk who come from one discipline and are increasingly looking at life sciences through the view of computing: The Biological Computation group is conducting research to uncover fundamental principles of biological computation: what cells compute, how and why. We focus primarily on developing computational techniques that enable multiscale modelling, from molecules to cells to systems. Our work currently focuses on fundamentals of Biological Computation, with applications in Immunology and Development, together with principles of Programming Life, with applications in DNA Computing and Synthetic Biology. Click on the image or title to learn more.
Scoop.it!
Having been tangentally* impacted by the Intel Larrabee project - which was to create a GPU based on the standard Intel architecture - it has been interesting following how intel is responding to the GPU adoption in the technical/ parallel computing market. It took fifteen years for Intel to shrink the computing power of the teraflops-busting ASCI Red massively parallel Pentium II supercomputer down to something that fits inside of a PCI-Express coprocessor card – and the Xeon Phi coprocessor is only the first step in a long journey with coprocessor sidekicks riding posse with CPUs in pursuit of exascale computing. With over 50 Pentium cores on the card - this promises to significantly impact the capabilities of desktop computing, server and cloud compute power. Between Nvidia's Kepler GPU and Intel's Xeon phi - its going to be an interesting 12 months- especially give that our new simulation tools are designed for this type of environment. To read the rest of the Register article click on the image or the title to learn more. ** ( re: tangental: they considered buying some software I was involved in bringing to market and when they didn't the team created a start up to commercialize it instead.)
Scoop.it!
Microsoft Research will present its Faculty Summit accessible by the public. On July 16 and 17, 9:00 a.m. to 2:00 p.m. Pacific Standard Time (PST) (noon to 5:00 p.m. Eastern Standard Time). The summit gives an opportunity to hear leading academic researchers and educators, as well as Microsoft researchers, product group engineers, and architects explore new opportunities in computer science research and advances that address real-world challenges. Should be at least and hopefully awesome. Click on the image or title to learn more.
|
NVIDIA is getting involved in standards for parallelization of compute at language level - this is likely a very good thing (and it won;t hurt them selling hardware either).