Practical network support for IP traceback
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
Congestion control for high bandwidth-delay product networks
Proceedings of the 2002 conference on Applications, technologies, architectures, and protocols for computer communications
User-level internet path diagnosis
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Source selectable path diversity via routing deflections
Proceedings of the 2006 conference on Applications, technologies, architectures, and protocols for computer communications
MIRO: multi-path interdomain routing
Proceedings of the 2006 conference on Applications, technologies, architectures, and protocols for computer communications
Multihoming performance benefits: an experiment evaluation of practical enterprise strategies
ATEC '04 Proceedings of the annual conference on USENIX Annual Technical Conference
Improving web availability for clients with MONET
NSDI'05 Proceedings of the 2nd conference on Symposium on Networked Systems Design & Implementation - Volume 2
NIRA: a new inter-domain routing architecture
IEEE/ACM Transactions on Networking (TON)
Proceedings of the ACM SIGCOMM 2008 conference on Data communication
X-trace: a pervasive network tracing framework
NSDI'07 Proceedings of the 4th USENIX conference on Networked systems design & implementation
OFRewind: enabling record and replay troubleshooting for networks
USENIXATC'11 Proceedings of the 2011 USENIX conference on USENIX annual technical conference
MAPLE: a scalable architecture for maintaining packet latency measurements
Proceedings of the 2012 ACM conference on Internet measurement conference
| Hi-index | 0.00 |
In this paper, we describe Net-Replay, a new network primitive to help application end points conduct in-band characterization of the glitches they encountered. In Net-Replay, each network infrastructure element remembers a small amount of information for every packet observed at the element over a certain time interval. Furthermore, network elements expose a simple "packet marking" interface, using which they can indicate to end-points whether or not they had seen a particular packet in the past. When application end-points observe glitches, they replay (i.e. retransmit) the packets which observed the glitch and leverage feedback from network elements to determine the type and location of the glitch encountered by the packets. We discuss how end-host network stacks should be modified to leverage Net-Replay in this fashion. We also consider how network infrastructure can support Net-Replay in a lowoverhead fashion. We argue that Net-Replay can enable applications to detect a variety of glitches and react to them in an accurate and informed manner, while ensuring that the infrastructure stays simple and fast. We believe that proactive support from the network in the form of Net-Replay-like functionality is crucial to ensure robust performance of future Internet applications, many of which are likely to be highly demanding and far less tolerant of network glitches than traditional applications.