HPC WORLD
High Performance Computing
|
Loading Digu...
|
Iranian Cyber Army Takes Over Baidu
Jan 13th
These days, the hottest news about Internet in China is that Baidu was attacked. Yanhong LI, the CEO of Baidu, even said it was unprecedented in history.
As reported, the most popular Web search engine in China, Baidu.com, was attacked by a group of hacktivists identifying themselves as the Iranian Cyber Army. The hackers hijacked the domain’s DNS records and pointed it to a server under their control.
Baidu is a Chinese search engine established in 2000. According to the Alexa traffic ranking, the website is currently number one in China and eight in the world. Additionally, Baidu.com dominates the search engine market in the country with over a 77 percent share.
The attack against the search engine occurred yesterday (January 12 2009), when for about three hours, its main page displayed the image of the Iranian flag and a message reading "THIS SITE HAS BEEN HACKED BY IRANIAN CYBER ARMY." Additional text written in Persian translates amongst the lines of "The Iranian Cyber Army has been established in protest to foreign countries and Zionists interfering in the domestic affairs of our country and broadcasting false news."
3D Microchips for More Powerful and Environmentally-Friendly Computers
Jan 12th
Years ago, home computers were powered by single-core processors, and in time, dual- and quad-core designs emerged which meant that two of four cores were placed alongside each other.
NOW, experts are looking to create 3D microprocessors, in which the cores will not be stacked next to each other, but on top of each other.
The vertical stacking allow for a much higher transfer speed between the cores – up to 10 times faster than currently possible – as well as for improved heat efficiency and overall performance.
But, efficiency is not the only reason why 3D architectures are researched. It is predicted that the supercomputers of 2100 would theoretically use up the whole of the US’ electrical supply. A new, revolutionary cooling system that would further reduce temperatures inside future 3D microprocessors is being studied now. The architecture of 3D microprocessor and its comparison with today’s multichip can be seen in the right picture.
BlueArc Storage Solution Powers Visual Effects Behind ‘Avatar’
Jan 11th
BlueArc Corporation is a network storage device manufacturer. Its products are network attached storage (NAS) systems.
The primary feature of the BlueArc’s NAS products is BlueArc’s heavy reliance on FPGA’s. The FPGA’s provide hardware acceleration of processing which typically occurs in software running on general purpose CPU’s on competing products. As a result, BlueArc is able to provide high performance for a single system or node. – For more details goes to Wiki/BlueArc
There is no doubt that the movie ‘Avatar’ is one of the hottest topics in the world these days. Putting aside the visual feast this film brings us, we should focus more on the wide range of the applications of high performance computing behind the ‘Avatar’.
As http://www.hpcwire.com reported on January 6 2010, Weta Digital uses BlueArc to manage unprecedented rendering workloads with high performance for ‘Avatar’. The data below is really prodigious:
Intel Ct Technology
Jan 10th
As was mentioned in <Intel Ct Language in HPC>, HERE is the post to discuss more about Ct.
On October 29, 2009, my former tutor with her students and I, attended a “Development of Parallel Computing Program and Advanced Debugging and Optimization” Training and Seminar, held by Intel at Shanghai Supercomputer Center (SSC). In the first lecture given by Joe Wolf, an expert of Intel Software & Services Group, a new concept named “Ct Technology” was proposed.
At that time, Joe raised a question that how many people presented there ever heard of “Ct” before, and nobody answered Yes. But Ct nowadays, is a very hot technology.
Ct = C for Throughput Computing
In the presentation, Joe illustrated some main features of Ct like:
• Ct adds parallel collection objects & methods to C++
– Works with standard C++ compilers (ICC, GCC, VC++)
• Ct abstracts away architectural details
– Vector ISA width / Core count / Memory model / Cache sizes
• Ct forward-scales software written today
– Ct is designed to be dynamically retargetable to SSE, AVX, Larrabee, and beyond
• Ct is deterministic*
– No data races
Vec is the most basic type of parallel collection object.
Here we give two codes to prove that Ct works with standard C++ compilers:
// C/C++ Code
#include "cpp_code.h"
int main()
{
// Dot Product Using C Loops
for (i = 0; i < n; i++) {
dst += src1[i] * src2[i];
}
}
#include "cpp_code.h"
int main()
{
// Dot Product Using C Loops
for (i = 0; i < n; i++) {
dst += src1[i] * src2[i];
}
}
// Ct technology Code
#include "ct.h"
int main()
{
// Dot Product Using Ct
Vec Src1(src1, n), Src2(src2, n);
F64 Dst = addReduce(Src1 * Src2);
}
#include "ct.h"
int main()
{
// Dot Product Using Ct
Vec Src1(src1, n), Src2(src2, n);
F64 Dst = addReduce(Src1 * Src2);
}
All the above were discussed on the seminar. And Next, we will provide more information about Ct which is available on Intel Website.
Intel Ct Language In HPC
Jan 7th
As Douglas Eadline (Ph.D) mentioned in From Ct to InfiniBand, The inside Scoop from SC09, there are two hardware trends in HPC. The first is multi-core computing. While this presents some issues, it still has the advantage that it provides some level of homogeneity i.e. all the cores are the same. The latest trend is heterogeneous computing where there are special purpose cores available for data parallel computations i.e. GP-GPU’s.
More details about Ct will be discussed in my later post.
Recent Comments