On network-aware clustering of Web clients
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
High Performance Single-Chip FPGA Rijndael Algorithm Implementations
CHES '01 Proceedings of the Third International Workshop on Cryptographic Hardware and Embedded Systems
TCP-Splitter: A TCP/IP Flow Monitor in Reconfigurable Hardware
HOTI '02 Proceedings of the 10th Symposium on High Performance Interconnects HOT Interconnects
A 21.54 Gbits/s Fully Pipelined AES Processor on FPGA
FCCM '04 Proceedings of the 12th Annual IEEE Symposium on Field-Programmable Custom Computing Machines
Computer Networks: The International Journal of Computer and Telecommunications Networking
Analysis of long duration traces
ACM SIGCOMM Computer Communication Review
Real-time anonymization in passive network monitoring
ICNS '07 Proceedings of the Third International Conference on Networking and Services
Legal issues surrounding monitoring during network research
Proceedings of the 7th ACM SIGCOMM conference on Internet measurement
Reconfigurable architecture for network flow analysis
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Packet-level traffic measurements from the Sprint IP backbone
IEEE Network: The Magazine of Global Internetworking
Relationships and data sanitization: a study in scarlet
Proceedings of the 2010 workshop on New security paradigms
Hi-index | 0.00 |
This paper describes a highly scalable architecture based on field-programmable gate-array (FPGA) technology for prefix-preserving anonymization of IP addresses at increasingly high network line rates. The Crypto-PAn technique, with the Advanced Encryption Standard (AES) as the underlying pseudo-random function, is fully mapped into reconfigurable hardware. A 32 Gb/s fully-pipelined AES engine was developed and used to prototype the Crypto-PAn architecture. The prototype was implemented on a Xilinx Virtex-4 device achieving a worst-case Ethernet throughput of 8 Gb/s using 141 block RAM's and 4262 logic cells. This is considerably faster than software implementations which generally achieve much less than 100 Mb/s throughput. A technology-independent analysis is presented to explore the scalability of the architecture to higher multi-gigabit line-rates.