EEDSP
17.7K views | +4 today
Follow
EEDSP
Digital Signal Processing, Data Analytics, Big Data, HPC, Deep Learning, GPGPU, Distributed and Parallel Computing
Curated by Shiwon Cho
Your new post is loading...
Your new post is loading...
Scooped by Shiwon Cho
Scoop.it!

Benchmarks: Intel Xeon Phi vs. NVIDIA Tesla GPU

Benchmarks: Intel Xeon Phi vs. NVIDIA Tesla GPU | EEDSP | Scoop.it

We’ve seen that there is one processor that needs to be added to the picture — the commodity multi-core CPU. This is already a part of many server configurations, and for some applications, e.g., Monte-Carlo pricing of American options, it can give better or comparable performance than an accelerator processor when optimized correctly. Between NVIDIA’s Kepler GPUs and Xeon Phi, the GPU wins for both of our test applications.

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

PNNL: Energy Star

PNNL: Energy Star | EEDSP | Scoop.it

To improve overall energy efficiency and performance on future exascale computing systems, scientists from PNNL; University of California, Riverside; and Marquette University examined some advanced high-performance computing systems and determined undervolting that also leverages existing mainstream resilience techniques at scale to reduce power consumption improved system failure rates. The undervolting method does not require modifying existing hardware or using pre-production machines and has shown positive results toward achieving a cost-efficient, energy-savings implementation for the HPC field. The paper documenting this first-of-its-kind work, Investigating the Interplay between Energy Efficiency and Resilience in High Performance Computing, will be presented during the IPDPS 2015 conferences technical program.

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

How League of Legends Scaled Chat to 70 million Players - It takes Lots of minions. - High Scalability -

How League of Legends Scaled Chat to 70 million Players - It takes Lots of minions. - High Scalability - | EEDSP | Scoop.it
How would you build a chat service that needed to handle 7.5 million concurrent players, 27 ...
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

IBM researchers make a chip full of artificial neurons

IBM researchers make a chip full of artificial neurons | EEDSP | Scoop.it

Coprocessors and a neural network supercomputer may follow. 

A team of scientists at Cornell University and IBM Research have gotten together to design a chip that's fundamentally different: an asynchronous collection of thousands of small processing cores, each capable of the erratic spikes of activity and complicated connections that are typical of neural behavior. When hosting a neural network, the chip is remarkably power efficient. And the researchers say their architecture can scale arbitrarily large, raising the prospect of a neural network supercomputer.

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Using SSD as a Foundation for New Generations of Flash Databases  - High Scalability -

Using SSD as a Foundation for New Generations of Flash Databases  - High Scalability - | EEDSP | Scoop.it
“You just can't have it all” is a phrase that most of us are accustomed to ...
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

The Best of Both Worlds: Low Cost and Large Scale OpenMP and MPI | Cluster Design | Features

The Best of Both Worlds: Low Cost and Large Scale OpenMP and MPI | Cluster Design | Features | EEDSP | Scoop.it
Cluster Monkey is an exclusive content based site that speaks directly to the high performance computing (HPC) cluster market and community. We focus on benchmarks, tutorials, case studies, and how-to information that is useful to cluster users, administrators, purchasers, and designers.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Paper: Can Programming Be Liberated From The Von Neumann Style?  - High Scalability -

Paper: Can Programming Be Liberated From The Von Neumann Style?  - High Scalability - | EEDSP | Scoop.it
Famous computer scientist John Backus , he's the B in BNF(Backus-Naur form) and the creator of...
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Colfax Research | Configuration and Benchmarks of Peer-to-Peer Communication over Gigabit Ethernet and InfiniBand in a Cluster with Intel Xeon Phi Coprocessors

Colfax Research | Configuration and Benchmarks of Peer-to-Peer Communication over Gigabit Ethernet and InfiniBand in a Cluster with Intel Xeon Phi Coprocessors | EEDSP | Scoop.it

Intel Xeon Phi coprocessors allow symmetric heterogeneous clustering models, in which MPI processes are run fully on coprocessors, as opposed to offload-based clustering. These symmetric models are attractive, because they allow effortless porting of CPU-based applications to clusters with manycore computing accelerators.

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Beagle Supercomputer is a Genome Smasher - insideHPC

Beagle Supercomputer is a Genome Smasher - insideHPC | EEDSP | Scoop.it
The Beagle Supercomputer at the University of Chicago can analyze 240 whole genomes in two days.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Scale your file system with Parallel NFS

Scale your file system with Parallel NFS | EEDSP | Scoop.it
The Network File System (NFS) is a stalwart component of most modern local area networks (LANs). But NFS is inadequate for the demanding input- and output-intensive applications commonly found in high-performance computing -- or, at least it was.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

OpenMP.org » OpenMP Specifications

OpenMP.org » OpenMP Specifications | EEDSP | Scoop.it

OpenMP 4.0 supports the programming of accelerators, SIMD programming, and better optimization using thread affinity

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

High Scalability - High Scalability - Strategy: Stop Using Linked-Lists

High Scalability - High Scalability - Strategy: Stop Using Linked-Lists | EEDSP | Scoop.it
What data structure is more sacred than the link list? If we get rid of it what silly interview...
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Ditching RAM may lead to low-cost supercomputers

Ditching RAM may lead to low-cost supercomputers | EEDSP | Scoop.it
Many servers, supercomputers and other monster systems thrive on high-speed RAM to keep things running smoothly, but this memory is wildly expensive -- and that...
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Hitachi have developed new CMOS semiconductor.

Hitachi have developed new CMOS semiconductor. | EEDSP | Scoop.it

Hitachi, announce CMOS semiconductor, a computer that is comparable to the quantum computer

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

HPCC Systems from LexisNexis Celebrates Third Open-Source Anniversary, Releases 5.0 Version - insideBIGDATA

HPCC Systems from LexisNexis Celebrates Third Open-Source Anniversary, Releases 5.0 Version - insideBIGDATA | EEDSP | Scoop.it
HPCC Systems from LexisNexis Celebrates Third Open-Source Anniversary, Releases 5.0 Version
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Shared Memory Clusters: Of NUMA And Cache Latencies

Shared Memory Clusters: Of NUMA And Cache Latencies | EEDSP | Scoop.it
In the introduction article in this series, I explained that NUMA topology-based SMP systems with two levels to the memory (DRAM) hierarchy have been aroun
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Flash Said to be More Cost-Effective than DRAM - Wikibon

Flash Said to be More Cost-Effective than DRAM - Wikibon | EEDSP | Scoop.it
This article explores how using flash as a persistent extension of main memory is a much more cost effective approach than purely relying on DRAM as an in-memory data store.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Scientific computing’s future: Can any coding language top a 1950s behemoth?

Cutting-edge research still universally involves Fortran; a trio of challengers wants in.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Managing GPUs by SLURM - insideHPC

In this video from the HPC Advisory Council Swiss Conference 2014, Massimo Benini from CSCS presents: Managing GPUs by SLURM.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Coding for SSDs – Part 1: Introduction and Table of Contents | Code Capsule

Introduction I want to make solid-state drives (SSDs) the optimal storage solution for my key-value store project. For that reason, I had to make sure I
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Intel® Xeon® Processor E7 V2 Family Technical Overview

Intel® Xeon® Processor E7 V2 Family Technical Overview | EEDSP | Scoop.it

The new Intel Xeon processor E7 v2 product family is designed to make data more valuable for your business through in-memory computing – one of the more recent advances in data management and analytic solutions, which stores the entire data set in main memory rather than traditional hard disk storage. In-memory database and analytics solutions enable significant performance gains in analyzing complex and diverse datasets. We’re talking about analysis in seconds or minutes rather than hours or days. This is how you get to real-time insight.

more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Networking scalability on high-performance servers

Networking scalability on high-performance servers | EEDSP | Scoop.it
The proliferation of high-performance scalable servers has added a new level of complexity to networking and system performance.
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Topic of Interest - Distributed Stream Processing

Topic of Interest - Distributed Stream Processing | EEDSP | Scoop.it
I've been interested in distributed stream processing since the Storm and S4 announcements. Storm was open sourced by Twitter and S4, now an Apache Incubator project, was open sourced by Yahoo! Now...
more...
No comment yet.
Scooped by Shiwon Cho
Scoop.it!

Creating parallel reactive and streaming applications with the Intel® Threading Building Blocks (Intel® TBB) flow graph | Intel® Developer Zone

Creating parallel reactive and streaming applications with the Intel® Threading Building Blocks (Intel® TBB) flow graph | Intel® Developer Zone | EEDSP | Scoop.it

The flow graph feature available in Intel® Threading Building Blocks (Intel® TBB) allows users to easily create both dependence graphs and reactive, messaging passing graphs that execute on top of Intel TBB tasks. Users programmatically create nodes and edges that express the computations performed by their application and the dependencies between these computations.

more...
No comment yet.