Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Packet detection on Maxeler architecture

Kapellaki Maria

Full record

---

Summary

Recent years, the Internet is a big part of our everyday life because access to it is very easy, through various devices. The need for effective protection and safety systems is necessary.Recently, Network Intrusion Detection Systems - NIDS, have been developed, often as an upgrade of firewalls; NID Systems not only check the headers of passing packet, but they also check the payload of the packet for the existence of specific suspicious strings. One of the most popular Network Intrusion Detection systems is \textbf{Snort}, which is a free and open source tool.We can easily understand that one string search in the data of a packet is not straightforward and can be very demanding in terms of time, and on system resources when we talk about high bit rates. A solution to this can be the use of hardware, as it can provide much higher processing rates than those of the software.This thesis is focused on the creation of such a system, on a hybrid supercomputer of the Maxeler Technologies Company, having SNORT as a reference system. Maxeler supercomputers, use an architecture that combines a general purpose processor with reconfigurable logic and leads to the construction of systems with excellent potentialities.In this thesis, we created a system for content matching, header matching, rule matching, as well as an automatic generator of code needed by Maxeler's compiler, thus composing a complete network intrusion detection system, following the principles and syntax of the Snort rules.

Available Files

Services