UniParthenope OpenSource Lab Party

https://i1.wp.com/osl.uniparthenope.it/wp-content/themes/coogee/images/uposl.pngOn the occasion of the new release of our main products, GVirtuS and JaClouX we are pleased to announce the first “UniParthenope OpenSource Lab Party”.

The party will be held on March 16 2011 in the Faculty of Science and Technology of the University of Napoli “Parthenope”.

The party will be organized as a conference where there will be held talks about the projects developed in the OpenSource Lab.

In particular will be presented:

  • GVirtuS, a framework for the implementation of split-driver model and the of virtulization of hardware devices.
  • JaClouX, an open source, cloud independent, Java API for high performance cloud computing application design, development, simulation and evaluation.

During the party it will be possible to meet and talk with the developers of the laboratory.

Further news about the party will be posted on the OSL website: http://osl.uniparthenope.it/tag/osl-party.

The Uniparthenope OpenSource Lab Team.

gVirtuS 01-beta2 released, supporting cudatoolkit 3.1 and 3.2

From this release (01-beta2) we support cudatoolkit >= 3.1. The legacy compatibility for cudatoolkit versions older than 3.1 is not guaranteed.

In this release there are also some minor changes and bugfixes.
The frontend now is named “libcudart.so.3”, so it is not needed anymore to rename it or to preload it.
The AfUnix communicator now has a new (optional) option, “mode” for setting the permission on the socket file (0660 for default).

Go to the Project Page and to the Download Page.

Writing CUDA applications using the D programming language

On the release note of Fedora 14 there is the introduction of the support for developing using the D programming language, so I’ve readed something on that language and I feel that probably I can actually use it, the main interesting feature for me is the simple interoperability with code written in C.

So just for testing purpose I decided to write a small CUDA (I’m pretty sure that you already know what is cuda) application using the D language.

Of course is not possible to write CUDA kernel and device functions directly in a D module so what is needed is to implement kernels in a cuda source file (with a proper launcher) and then use the kernel from the D module.

The application that i’ve written converts a string in uppercase, reading the input from the command line (remember is just for testing).
Read more of this post

I'm the science (aka my first publication)

In these last days my first paper had been published. The title of the paper is “A GPGPU Transparent Virtualization Component for High Performance Computing Clouds”. It’s an article about gVirtuS: the implementation, the usage and a comparison with other software that aim to accomplish to do the same job (seriously in the real world there isn’t any competitor).

I wrote this article with the staff of the Applied High-Performance Scientific Computing Research Laboratory of the “Department of Applied Science” (DSA)University of Napoli “Parthenope”. This work was presented at Euro-Par 2010.

For the occasion I’ve added a new page to my website to collect my Publications (yes, I hope that I’ll write some other papers).

gVirtuS: the first beta release

We are proud to announce the first beta release of gVirtuS.

gVirtuS allows an instanced virtual machine to access GPGPUs in a transparent way, with an overhead  slightly greater than a real machine/GPGPU setup. gVirtuS is hypervisor independent, and, even though it currently virtualizes nVIDIA CUDA based GPUs.

The software, developed for research applications, is provided as it is.

We encourage using and testing it in order to collect useful feedbacks and suggestions.

Take a look to the gVirtuS project page: http://osl.uniparthenope.it/projects/gvirtus/.

Using Nvidia drivers on Fedora 12 with DKMS

Although it sucks too much using closed source software, someone wants or have to use that. It’s the case of Nvidia graphics drivers: someone wants them to obtain better performances, others have to use them because their adapters don’t work well with open source drivers, and others (like me) have to use Nvidia closed drivers for developing and using CUDA applications.

This is essentially a manual installation but thanks to DKMS, Dynamic Kernel Module Support, it isn’t required anymore to reinstall the driver at every kernel update.

Read more of this post