Home
Improving the performance of network intrusion detection systems using graphics processors
Results - Details
|
||||
| Identifier | 000339004 | |||
| Title | Improving the performance of network intrusion detection systems using graphics processors | |||
| Alternative Title | Βελτίωση της Απόδοσης των Συστημάτων Ανίχνευσης Επιθέσεων Δικτύων χρησιμοποιώντας Επεξεργαστές Γραφικών | |||
| Author | Βασιλειάδης, Γεώργιος Ιωάννη | |||
| Thesis advisor | Μαρκάτος, Ευάγγελος | |||
| Abstract |
In this thesis, we present an intrusion detection system based on the Snort open-source NIDS that exploits the underutilized computational power of modern graphics cards to offload the costly pattern matching operations from the CPU, and thus increase the overall processing throughput. Graphic cards processing units (GPUs) provide high parallelism and offer significant computational power. SPMD (Single Process, Multiple Data) operation is ideal for pattern matching purposes since we can search in parallel for multiple patterns over multiple packets. We ported single and multi-pattern matching algorithms to the GPU and compared their performance with the corresponding CPU implementations. Our prototype system, called Gnort, achieved a maximum traffic processing throughput of 2.3 Gbit/s using synthetic network traces, while when monitoring real traffic using a commodity Ethernet interface, it outperformed unmodified Snort by a factor of two. The results suggest that modern graphics cards can be used effectively to speed up intrusion detection systems, as well as other systems that involve pattern matching operations up to 8x times faster. |
|||
| Physical description | xviii, 51 σ. : εικ. ; 30 cm. | |||
| Language | English | |||
| Issue date | 2008-12-04 | |||
| Collection | School/Department--School of Sciences and Engineering--Department of Computer Science--Post-graduate theses | |||
| Type of Work--Post-graduate theses | ||||
| Views | 536 | |||
| Digital Documents | |
|---|---|
|
Download document |
