Tuesday, December 21, 2010

GPU & Parallel Computing Workshop


GPU & Parallel Computing Workshop
January 6th 2011

SagivTech and Microsoft R&D Israel are happy to invite you to a workshop on GPU & Parallel Computing.

GPU and parallel computing provide developers a great means for acceleration and speed up of computationally intensive applications.

In this workshop some recent work conducted in the Israeli industry and academy will be presented. Moreover, insights regarding recent international conferences in these fields will be given.

The workshop will be conducted at Microsoft R&D, Shenkar 13, Herzliya, on Thursday January 6th, 14-18.

This workshop is free of charge, but pre-registration is required.

For more details contact: chen@sagivtech.com
For registration visit: http://www.sagivtech.com/21262.html

Dr. Avi Mendelson, MSFT
Dr. Chen Sagiv, SagivTech
Latest trends in GPU Computing – impressions from GTC 2010
Micha Feigin, SagivTech
GPU for Industrial Inspection Applications
Gonen Rave, Orbotech
GPU@HP Research
Dr. Sagi Schein, HP Research
Multi core DSPs vs GPUs
Nissim Saban, Texas Instruments
Coffee Break
Parallel Genetic Algorithms on the GPU
Dr. Alon Kongreen, Bar Ilan University
Why use GPU ?
Eri Rubin, OptiTex
Impressions from Super Computing 2010
Dr. Guy Tel Zur, Ben Gurion University and GRID
Compute API – past & future
Ofer Rozenberg, Intel
End of Moore’s law or end of General Purpose Computing?
Dr. Avi Mendelson, MSFT
Wrap Up

Chen Sagiv, PhD
SagivTech Ltd.
Mobile: +972-54-7706089
Fix: +972-9-7411561

Thursday, November 11, 2010

IGT HPC Work Group Meeting

When: Wednesday, November 24rd, 2010

Where: IGT Offices, Maskit 4, 5th Floor, Hertzliya.

13:45-14:00: Gathering
14:00-14:10: Opening - Avner & Guy
14:10-15:00: Geomage GPU Cluster - Case Study
15:00-15:15: Coffee Break
15:15-16:00: Impressions from Super Computing 2010 Conference
16:00-16:10: Discussion and concluding remarks
Detailed Agenda:

14:10-15:00:  Geomage GPU Cluster - Case Study
Speaker: Eyal Hirsch, Senior HPC developer, Geomage

Geomage has been using Tesla line GPUs in production for the last 2 years. The cluster operates at high volumes, in terms of computation power, processing seismic projects around the clock.
Presentation will talk about seismic processing and the issues we've encountered from the development stage, QA, multi-GPU all the way to production including a real world case study.

15:00-15:15: Coffee Break

15:15-16:00: Impressions from Super Computing 2010 Conference
Speaker: Dr. Guy Tel-Zur – IGT Chairman.

The SC, Super Computing, conference is the World’s leading event in High-End Computing. The talk will cover Super Computing 2010 that will take place this November in New Orleans. Major announcements, including the latest Top500 Super Computing ranking, new products and trends will be discussed.

16:00-16:15: Discussion and concluding remarks

To register, please send your contact details to info@cloud.org.il
We are looking forward to seeing you!
Best Regards,

Guy Tel-Zur, Ph.D.
IGT Chairman

Date Nov 24, 2010 14:00 16:00
Location IGT Offices

Monday, August 23, 2010

Testing Intel’s Parallel Studio

Part 1 – OpenMP

Intel claims to bring simplified “end-to-end parallelism to Microsoft Visual Studio C/C++ developers with Intel® Parallel Studio” [1].

I used a simple OpenMP parallel HelloWorld program to study this new tool which comes as an add-on to Visual Studio 2008.The program listing is enclosed in Figure 1.

   1:  #include <omp.h>
   2:  #include <stdio.h>
   3:  // a function to consume cpu time:
   4:  void consume() {
   5:      int i;
   6:      long n=100000000;
   7:      double s=0;
   8:      for (i=1;i<n;i++)
   9:          s+=(double)1/(double)i;
  10:  }
  11:   int main (int argc, char *argv[]) {
  12:     int th_id, nthreads=5;
  13:     omp_set_num_threads(nthreads);
  14:     #pragma omp parallel private(th_id)
  15:     {
  16:       th_id = omp_get_thread_num();
  17:       consume();
  18:       printf("Hello World from thread %d\n", th_id);
  19:       #pragma omp barrier
  20:       if ( th_id == 0 ) {
  21:         nthreads = omp_get_num_threads();
  22:         printf("There are %d threads\n",nthreads);
  23:       }
  24:     }
  25:     return 0;
  26:   }

Figure 1: A “HelloWorld” in C with OpenMP directives

After compiling the code I tested it with the Parallel Amplifier (Profiler).

My machine has two physical cores and since it uses hyper-threading it looks as if it has 4 cores. I executed the program with 4 threads. In Figure 2 a screen dump of the profiler is shown.


Figure 2: Visual Studio 2008 with Parallel Studio screen dump

Let’s zoom at lower right plot:


Figure 3: A summary plot of the CPU resources utilization

In Figure 3 we can see that there were 4 threads on 4 cores and the performance was  optimal.

I would however recommend everyone to repeat every execution several times since there is a large variability between runs and taking average  results might be a good idea.

Focusing now on the center of the Visual Studio screen, Figure 4,


Figure 4: The Intel Parallel Studio Profiler

one can see the actual profiler analysis which points to the bottleneck in the code, i.e. the function consume() in this case.

Of course there is a lot to go deeper but as a first impression Parallel Amplifier can be helpful for OpenMP applications (and TBB too). Now let’s see what about MPI.

Part 2 – MPI

Originally MPI was created for Distributed Memory systems but MPI can also be used on a single (SMP) machine, like OpenMP, by using multiple threads. I installed both MPICH2 and Microsoft HPC Pack 2008 SDK on my computer.

Both can be set inside Visual Studio [2]. MPI programs, by default, are being executed using mpiexec executable.

There is no problem to compile the code with Visual Studio and to run it from the command line, but this way Parallel Studio won’t be useful. Therefore, one must execute the program from inside Visual Studio. This is doable. One should insert the mpiexec and the number of processes, e.g. –n 4 to the right place in the project properties, see Figure 5.

imageFigure 5: Setting mpiexec and the number of tasks from inside Visual Studio.

But here is the catch: since mpiexec is already compiled, its source is not given and in addition it is not part of the built (project), Parallel Amplifier can not analyze the user program and it only “sees” the mpiexec and therefore it only reports about mpiexec cpu consumption, see Figure 6.


Figure 6: The profiler can not penetrate beyond mpiexec and dive into the user’s code.

Therefore unless mpiexec will somehow be skipped by Parallel Studio it looks that this tool can not help much to analyze MPI applications!

You are welcome to post your comments / corrections.


[1] Intel Parallel Studio home page: http://software.intel.com/en-us/intel-parallel-studio-home/

[2] http://blogs.msdn.com/b/risman/archive/2009/01/04/ms-mpi-with-visual-studio-2008.aspx

Monday, August 02, 2010

Nvidia Nsight Installation Failed

I have a Windows 7 (64bits) OS with Visual Studio 2008 with SP1.
After starting  Parallel_Nsight_Host_Win64_1.0.10200 (Jul 2010) I got the following error message: “The required version of Visual Studio (2008 SP1) was not detected” – see screen capture:
I guess the problem is in the system registry of VS2008 which partly located  as a 32bit system and partly as a 64bit system and Nsight didn’t find what it expected.

 SP1 seems to be installed as can be seen in the image below:
It turned out that although SP1 is installed, it was not installed correctly.
After trying to reinstall SP1 I got the following error message:

I checked the log file:
Exe (C:\Users\telzur\AppData\Local\Temp\Microsoft Visual Studio 2008 SP1\VC_x86Runtime.exe) succeeded.[8/6/2010, 18:34:11] (IronSpigot::ExeInstallerBase::PerformAction) Log File: dd_VC_x86RuntimeMSI18CE.txt[8/6/2010, 18:34:11] (IronSpigot::ExeInstallerBase::PerformAction) Log File: dd_VC_x86RuntimeUI18CE.txt
[8/6/2010, 18:34:11] Action complete
[8/6/2010, 18:34:11] (IronSpigot::RegKeyT::Evaluate) RegKey: HKLM\SOFTWARE\Microsoft\VisualStudio\9.0 exists.
[8/6/2010, 18:34:11] Exists evaluated to true
[8/6/2010, 18:34:11] (IronSpigot::ExeInstallerBase::ExeInstallerBase) Created new ExePerformer for Exe item[8/6/2010, 18:34:11] (IronSpigot::CartmanExeInstallerBase::CartmanExeInstallerBase) Created new CartmanExePerformer for Exe item[8/6/2010, 18:34:11] (IronSpigot::CartmanExeInstaller::CartmanExeInstaller) In CartmanExeInstaller::CartmanExeInstaller[8/6/2010, 18:34:11] (IronSpigot::CompositePerformerBaseT >::PerformAction) Created new ExePerformer for Exe item
[8/6/2010, 18:34:11] (IronSpigot::CartmanExeInstallerBase::PreCreateProcess) In PreCreateProcess [8/6/2010, 18:34:11] (IronSpigot::CartmanExeInstallerBase::PreCreateProcess) m_pSetupWatcher->Connect succeeded [8/6/2010, 18:34:11] (IronSpigot::ExeInstallerBase::Launch) Launching CreateProcess with command line = VC_x64Runtime.exe /q /progress CartmanSetupExeWatcher3841249730 [8/6/2010, 18:34:12] (IronSpigot::CartmanExeInstallerBase::PostCreateProcess) PostCreateProcess succeeded [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::PerformAction) Exe log file(s) : [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::PerformAction) C:\Users\telzur\AppData\Local\Temp\dd_VC_x64RuntimeMSI18FC.txt [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::PerformAction) C:\Users\telzur\AppData\Local\Temp\dd_VC_x64RuntimeUI18FC.txt [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::PerformAction) Exe (C:\Users\telzur\AppData\Local\Temp\Microsoft Visual Studio 2008 SP1\VC_x64Runtime.exe) succeeded. [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::PerformAction) Log File: dd_VC_x64RuntimeMSI18FC.txt [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::PerformAction) Log File: dd_VC_x64RuntimeUI18FC.txt
[8/6/2010, 18:34:22] Action complete
[8/6/2010, 18:34:22] (IronSpigot::RegKeyT::Evaluate) RegKey: HKLM\SOFTWARE\Microsoft\VisualStudio\9.0 exists.
[8/6/2010, 18:34:22] Exists evaluated to true
[8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::ExeInstallerBase) Created new ExePerformer for Exe item[8/6/2010, 18:34:22] (IronSpigot::CartmanExeInstallerBase::CartmanExeInstallerBase) Created new CartmanExePerformer for Exe item[8/6/2010, 18:34:22] (IronSpigot::CartmanExeInstaller::CartmanExeInstaller) In CartmanExeInstaller::CartmanExeInstaller[8/6/2010, 18:34:22] (IronSpigot::CompositePerformerBaseT >::PerformAction) Created new ExePerformer for Exe item
[8/6/2010, 18:34:22] (IronSpigot::CartmanExeInstallerBase::PreCreateProcess) In PreCreateProcess [8/6/2010, 18:34:22] (IronSpigot::CartmanExeInstallerBase::PreCreateProcess) m_pSetupWatcher->Connect succeeded [8/6/2010, 18:34:22] (IronSpigot::ExeInstallerBase::Launch) Launching CreateProcess with command line = VC_IA64Runtime.exe /q /progress CartmanSetupExeWatcher2516872804 [8/6/2010, 18:34:22] (IronSpigot::CartmanExeInstallerBase::PostCreateProcess) PostCreateProcess succeeded [8/6/2010, 18:34:31] (IronSpigot::ExeInstallerBase::PerformAction) C:\Users\telzur\AppData\Local\Temp\Microsoft Visual Studio 2008 SP1\VC_IA64Runtime.exe - Exe installer's log file/hint (%temp%\dd_VC_IA64Runtime*.txt|%temp%\..\dd_VC_IA64Runtime*.txt) does not exist or is invalid. [8/6/2010, 18:34:31] (IronSpigot::ExeInstallerBase::PerformAction) Exe (C:\Users\telzur\AppData\Local\Temp\Microsoft Visual Studio 2008 SP1\VC_IA64Runtime.exe) failed with 0x80070643 - Fatal error during installation. . [8/6/2010, 18:34:31] (IronSpigot::ExeInstallerBase::PerformAction) PerformOperation on exe returned exit code 1603 (translates to HRESULT = 0x80070643)
[8/6/2010, 18:34:31] Action complete
[8/6/2010, 18:34:31] (IronSpigot::LogUtils::LogFinalResult) Final Result: Installation failed with error code: (0x80070643), Fatal error during installation.

Then I searched the internet for similar error reports. Thanks to this post from Pieter's Blog I decided to follow his advice and to uninstall VS2008 Remote Debugger, then I reinstalled SP1 and this time with success!
This allowed me to finally install Nvidia Nsight:

Saturday, July 31, 2010

Sharing USB disk with Virtual Box guest OS

Assuming the host OS is Windows and the guest OS is Linux (CentOS in my case):
make a directory at the guest OS for the flash storage:
mkdir /home/telzur/shared
Plug the memory stick and check the guest OS log:
tail /var/log/messages
find the name of the new device, in my case, ./dev/sdb1
as root:
mount –t vfat /dev/sdb1 /home/telzur/shared
Your files are now in the shared folder.

Thursday, July 22, 2010

Parallel Computing with Python and MPI

I installed PyMPI (version 2.5b0) on my CentOS 5.5 (as a VirtualBox guest system).
The following short demo works:
import mpi
import time
print "Hello World! my rank=", mpi.rank, " group size is=", mpi.size
see screen capture:

Before installing PyMPI I installed these packages:
2. Python-develop

Installation of PyMPI:
1. ./configure --with-includes=-I/usr/local/include
2. make
3. make check
4. make install  (as root)

Good luck!

Saturday, July 03, 2010

A day with my laptop

When I came in the morning to work on my laptop I discovered that it had restarted due to some critical updates by Microsoft. One hour later iTunes opened a pop-up window and asked me to install a new version. So I allowed it. Tons of megabytes were downloaded and network became slow. Then, The AVG Anti-Virus had updated its Definition Table. I like when this happens, it gives me a good feeling that my laptop is well taken care, but then also appeared a notification about a serious patch which required a restart. OK, I confirmed that. At noon, Firefox installed a new version and requested a permission to restart as well, So I confirmed that too. Then, a few of my favorites plug-ins had stopped working because they were not compatible anymore with the new version of Firefox. Just then, without any correlation, the USB driver hung and the mouse was dead. so I gave up the mouse and continued to work with the touch pad. Then, the computer was slow. I got angry and opened the Task Manager to see what is going on. Firefox consumed 800MB and tens of other processes filled the whole memory, so I started to kill a few. My ThinkPad is full of utilities by Lenovo. Nevertheless several processes were restarted because other mysterious processes keep an eye on them. Then, explorer died. OK, I am used to that so I restarted it manually. Then, I was told that there are unused icons on my desktop. Whenever I see this message I know that something really bad is going to happen. Never follow this advice. So I ignored it. OK, now everything seemed to be stable, I went to bring coffee, but when I returned to my computer there was a Javascript error and the browser asked me if I want to wait more or to stop it. Needless to say, I didn't know what to choose so I randomly selected to kill it. Silence. Firefox was Not Responding. I thought to myself that at this point nothing could be better than to restart the machine again. So I restarted to computer. Ten more minutes were lost while it was trying to boot. Finally it came back. I happily opened Word and started to work. When I wanted to print my document nothing happened. The printer spooler had died. Finally I got the point; I need to buy a new laptop.

Saturday, May 22, 2010

How to count the number of cores in a Linux machine

Use the following command:

cat /proc/cpuinfo | grep processor | wc –l

It is convenient to put it in a script which I will call, for example, ncores:

cat /proc/cpuinfo | grep processor | wc –l

See enclosed screen shot:


Saturday, May 08, 2010

Firefox 3.6 beta is still buggy

My Firefox version 3.6.4 is eating a lot of memory and the longer it is kept open the more memory it consumes, probably a memory leak somewhere. It seems that the blame is on the Plug-ins with their buggy JavaScript codes. Any how the user experience is not as expected.
Enclosed a screen dump of the task manager after killing Firefox plugin-container.exe
* This post was prepared using Microsoft’s Windows Live Writer.

Thursday, April 29, 2010

IGT Event: Advanced Cloud Based Solutions

Advanced Cloud based solutions:
Voltaire-IGT hosting eBay, XIV in Voltaire Cloud Centre of Excellence

May 12, 09:30-13:00
Location : Voltaire Offices, 13 Zarchin St., Raanana

The agenda:
09:30 -10:00 Coffee and croissant gathering
                                              Avner Algom: IGT Welcome
10:00-10:30  “ Cloud Centre of Excellence
                                              Asaf Somekh, Voltaire
10:30-11:00    "Grid-based storage: The complementary solution for cloud computing"
Hanoch Ben-David, XIV
11:00-11:30 “ Virtualized Data Centre
                                            Yaron Haviv, Voltaire  
11:30-12:00 “ Cloud Infrastructure Management – UFM demo
                                            Tom Thirer, Voltaire  
12:00-12:30 " eBay Case Study "
  Cees de Groot, eBAY, Head of Advanced Technology of  eBay's leading classifieds site in the Netherlands
Open Source advocate at The Net
12:30-13:00 Visit to IGT- Voltaire Cloud Lab
The IGT Cloud Lab with Voltaire's Cloud Centre of Excellence
provides the IGT members with innovative platforms to R&D cloud based solutions.
To reserve your place, please send the event name with your contact details to: info@grid.org.il

Friday, March 05, 2010

IGT HPC WG Meeting - High performance computing needs for Numerical Weather Prediction

IGT HPC Work Group Meeting

High-performance computing needs for Numerical Weather Prediction

Speaker: Yoav Levi, Israel Meteorological Servicehttp://www.ims.gov.il/IMS/All_Tahazit/homepage.htm
Wednesday, March 24th, 2010IGT Offices, Maskit 4, 5th Floor, Hertzliya.
13:45-14:00: Gathering
14:00-14:05: Opening - Avner & Guy
14:05-15:00: High performance computing needs for
                      Numerical Weather Prediction
15:00-15:15: Discussion and concluding remarks
High performance computing needs for Numerical Weather Prediction
Abstract:Numerical Weather Prediction (NWP) uses current weather conditions as input into Mathematical models of the atmosphere to predict the weather. Richardson wrote the differential equations for NWP at 1922. He also calculated that if there was a "forecasting factory", 64,000 people will be needed to solve the equations and produce a forecast for the next day.
It wasn't until the advent of the computer and computer simulation that it was feasible to do NWP in real-time. Manipulating the huge datasets and performing the complex calculations necessary to do this on a resolution fine enough to make the results useful requires the use of some of the most powerful supercomputers in the world.
At 1963 Lorenz discovered that small changes in initial conditions produced large changes in the long-term outcome. Lorenz's discovery was the base of the Chaos theory and proved NWP could not reasonably predict weather beyond a weekly period. The development of parallel computing opened the opportunity to run numerous runs with different initial conditions. These ensemble models define the forecast uncertainty and extend weather forecasting farther into the future than would otherwise be possible.
To register, please send your contact details to info@grid.org.il
We are looking forward to seeing you!
Guy Tel-Zur, Ph.D. IGT HPC WG Head & Chairman 

Tuesday, February 23, 2010

IGT HPC Work Group Meeting - Scalable x86-based solutions for HPC

IGT HPC Work Group Meeting - Scalable x86-based solutions for HPC Users: Clusters, Cloud, and Shared Memory (SMP) in a unified platform

Wednesday, March 3rd, 2010
IGT Offices, Maskit 4, 5th Floor, Hertzliya.
13:45-14:00: Gathering
14:00-14:05: Opening - Avner & Guy
14:05-15:00: Part 1
15:00-15:10: Coffee Break
15:10-16:00: Part 2
16:00-16:15: Discussion and concluding remarks

Title:Scalable x86-based solutions for HPC Users: Clusters, Cloud, and Shared Memory (SMP) in a unified platform
Abstracts:IBM (Gilad Berman, System x Technical Specialist)Scalable Data Center alternatives: iDataplex, BladeCenter, new management softwareiDataPlex technology offers a new innovative design to address the needs of the data center with extreme density and simplified manageability while reducing power and cooling consumption.BladeCenter is an established, open design with a flexible, scalable architecture that offers a true alternative to today’s sprawling racks and overheated server rooms. By integrating servers, storage, networking and management, BladeCenter is helping companies maximize performance, minimize costs and sweep complexity aside.xCAT is a new advanced generation of cluster deployment and management system.
ScaleMP (Benzi Galili, Executive Vice President)SMP machines (shared memory, symmetric multi-processor) are still required for several classes of HPC workloads, mainly for use of very large (>0.5TB) RAM and also for high-core counts. In the past, building SMPs required tens to hundreds of millions of ongoing project investments for custom-chipsets, and in many cases involved custom-everything (CPUs, interconnect, chipset, OS, and software stacks) – making this type of technology cost-prohibitive for many science and engineering organizations. Past commercial attempts at creating software-based SMP (a.k.a Software-DSM) have failed. ScaleMP uses a novel virtualization-based approach, taking advantage of high bandwidth commodity interconnect and high-performance standard (x86) processors, to create SMPs that perform better for HPC workloads, and cost significantly less than the hardware-only alternatives. This session will describe the types of workloads that benefit from the architecture (including use cases around the world), and some of the basic ideas behind the architecture.
Three usage classes will be discussed: classic SMP, HPC cluster simplification, and Cloud/Grid oriented SMP “on-demand” services.
16:25-16:35: Discussion and concluding remarks
 To register, please send your contact details to: info@grid.org.il

We are looking forward to seeing you!
Guy Tel-Zur, Ph.D. IGT Chairmanwww.grid.org.il

Thursday, February 04, 2010

CBSA: Cloud-based Services and Applications

IEEE 2010

CBSA: Cloud-based Services and Applications

Miami Florida, USA
Date: July 5 - 10, 2010

Workshop chair

Prof. Mark Baker, SSE, University of Reading, UK.
Email: mark DOT baker AT computer DOT org


Cloud computing is increasingly being used for what was known as "on-demand" and "utility computing". The services provided, the APIs and the applications that can be hosted by these Cloud providers have superseded the use of the grid, and are increasingly becoming popular with users. The Cloud is a paradigm shift, where details are abstracted from the users; they no longer need expert knowledge about using the various technologies and running their applications on service infrastructure. Cloud computing provides a consumption and delivery model for IT services based on Internet, and it typically involves the provision of dynamically scalable and virtualized services.

There services that are provided by various Cloud providers, which include commercial entities, such as Amazon and Google, as well as those that are open-source systems, such as provided by Open Cirrus and Eucalyptus.

In this workshop we will examine and explore the services, interfaces and types of applications that can be executed on Cloud systems. In addition, we are interested in the interfaces used to access the underlying services, the pros¢ons of using virtualization, the range and scope of applications that can be executed, the security used by these services, and aspects such a service level agreements and quality of service provided. In addition, we are keen to know about the services that are used, and whether these are appropriate for various types of applications, such as sophisticated workflows and HPC.

Submission Dates

* April 1st, 2010, paper submission,
* April 15th, 2010, notification,
* April 30th, 2010, camera-ready version due.

List of topics

* Security Policies and Mechanisms,
* Possession and ownership of data,
* Scheduling on Clouds,
* Cloud benchmarks,
* Cloud-based services and protocols,
* Cloud Interoperability,
* Cloud Frameworks and Services,
* Storage and file systems,
* Cloud scalability and performance,
* Fault-tolerance,
* Application development and debugging tools,
* Business models and economics of cloud services,
* Vituralization of hardware and software resources,
* Performance monitoring and prediction of cloud performance,
* Capacity planning and resource allocation,
* Service Level Agreements and Quality of Services,
* Cloud architectures,
* Use of various Applications on Cloud-based systems.

Programme Committee

* Mark Baker, SSE, University of Reading, UK,
* Lee Gillam, University of Surrey, UK,
* Dejan Milojicic, HP, USA,
* Garry Smith, SSE, University of Reading, UK<
* Richard Sinnot, National e-Science Centre, University of Glasgow, UK,
* David Walker, Computer Science, Cardiff University, UK,
* Dave de Roure, ECS, University of Southampton, UK,
* Ken Hawick, Computer Science, Massey University, NZ,
* Geoffrey Fox, Indiana University, USA,
* Guy Tel-Zur, Israeli Association of Grid Technologies,
* Ewa Deelman, USC Information Sciences Institute, USA,
* Jeremy Cohen, Imperial College, UK,
* Terry Harmer, Belfast e-Science (BeSC), UK,
* Daniel S. Katz, University of Chicago, USA,
* David Wallom, Oxford e-Research Centre, UK,
* Marcel Kunze, Karlsruhe Institute of Technology (KIT), Germany,
* Andrew L Wendelborn, University of Adelaide, Australia,
* Nikolaos Antonopoulos, University of Derby, UK,
* James Farnhill, JISC UK,
* Matthew Dovey, JISC, UK,
* David Bacigalupo, ECS, University of Southampton, UK,
* Xiaoyu Chen, ECS, University of Southampton, UK.  


IEEE2010: http://www.servicescongress.org/2010/
CBSA: Cloud-based Services and Applications: http://acet.rdg.ac.uk/events/index.php

Feb-16-2010, URL Update:  Complex HPC

Thursday, January 07, 2010

Tuesday, January 05, 2010

The "Never Say Never" rule of Cloud Computing

"To any computing resource there might be a possible demand that will break it"

Saturday, January 02, 2010

Teaching CUDA/OpenCL in my course

Starting from Semester 2010B I am going to incorporate CUDA/OpenCL Programming for GPGPUs as a chapter in my Introduction to Parallel Processing course ( link to 2010A course).