High-performance networking

In the context of activities on ANTS (The Testbed on Autonomic NeTworkS) of the Comics research group, we are working on multi-core architectures suitable for high speed networking, such as Network Processors and GPGPUs.

---

Network Processors

Intel Corporation kindly donated four network processor boards to University of Napoli:

• Two of these boards are equipped with Intel IXP425@533MHz, 256MB SDRAM, two Intel PRO 2200 Wireless and two Ethernet cards.
• The other two boards are equipped with Intel IXP2350@900MHz, 512MB DDR RAM with ECC, two Gigabit Ethernet and one Fast Ethernet cards.

The donation represents an important contribution to our research activities. The University of Napoli deeply appreciated Intel Corporation for its valuable support.

Papers

A. Botta, W. de Donato, A. Pescapè and G. Ventre, "Networked Embedded Systems: a Quantitative Performance Comparison", Globecom 2008, New Orleans, LA, USA, December, 2008

Technical Reports

Alessio Botta, Antonio Pescapè,"How to set up and configure StarEast in Access Point Mode with Intel 2200 Wireless Network Interfaces", TR-DIS-NP-2-2007, Technical Report Dipartimento di Informatica e Sistemistica, Università di Napoli "Federico II".

Student Contributors

- Giuseppe Bruno
- Vincenzo Siressi
- Francesco Rendine

---

GPGPUs

We equipped a high-end server with an NVIDIA Tesla C1060 card (240 CUDA cores, 4GB GDDR3 memory), and an NVIDIA Tesla C2075 card (448 CUDA cores, 6GB GDDR5 memory), kindly donated to us by NVIDIA, in order to exploit the CUDA framework for implementing high-speed complex packet processing. In details, we are currently exploiting CUDA to implement efficient traffic classfication techniques to be used into the TIE platform.

Papers & Technical Reports

available soon (we hope...)

Student Contributors

- Stelvio Troncone

---

If you are interested in collaborating with us or in opportunities in Traffic, please send an e-mail to Antonio Pescapè