Increasing TCP Throughput with an Enhanced Internet Control Plane

@article{Bavier2006IncreasingTT,
  title={Increasing TCP Throughput with an Enhanced Internet Control Plane},
  author={Andy C. Bavier and Larry L. Peterson and Jack Brassil and Rick McGeer and David P. Reed and P. Sharma and Praveen Yalagandula and A. Henderson and Larry Roberts and Stephen Schwab and R. Thomas and E. Wu and Brian Princeton Mark and B. Zhao and A. Joseph},
  journal={MILCOM 2006 - 2006 IEEE Military Communications conference},
  year={2006},
  pages={1-7}
}
CHART seeks to improve the performance of operational DoD Internets through the introduction of an intelligent network overlay. TCP performance-particularly between CONUS and forward-deployed components located in combat theaters-can be severely degraded due to high loss rates and long latencies. The lack of current information about network conditions in the core NIPR-net/SIPRnet further compounds the problem, because end hosts lack the data required to make intelligent routing decisions… Expand

Figures from this paper

A Rate-based Congestion Control Overlay System
TLDR
A bandwidth probe control (BPC) system is designed for an overlay link, which estimates the delay and loss characteristics of the overlay link by probing and then uses this information dynamically to determine an appropriate transmission rate for the link. Expand
The Case for Service Overlays
  • Jack Brassil, R. McGeer, +4 authors S. Schwab
  • Computer Science
  • 2009 Proceedings of 18th International Conference on Computer Communications and Networks
  • 2009
TLDR
This paper introduces the concept of a Service Overlay network, to offer circuit-switched behavior on legacy IP networks, and establishes the requirements on the underlying IP network to make this strategy effective. Expand
Dynamic TCP Proxies: Coping with Disadvantaged Hosts in MANETs
TLDR
By enclosing poor quality path segments between a pair of TCP proxies, this work isolates the effect of these segments on the rest of the network and improves the overall performance. Expand
Performance of an Adaptive Routing Overlay under Dynamic Link Impairments
TLDR
An adaptive software routing overlay adaptively distributes traffic over a set of alternative paths based on real-time link status information provided by a system of network sensors to improve the performance of TCP/IP-based internets over links with dynamic impairments. Expand
A Multipath Flow Routing Approach for Increasing Throughput in the Internet
  • B. Mark, S. Zhang
  • Computer Science
  • 2007 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing
  • 2007
TLDR
A multipath flow routing overlay system, coupled with network sensors that provide realtime information about overlay links, to distribute traffic more efficiently over the network to improve overall throughput and network utilization is proposed. Expand
A Rate-based Congestion Control Overlay System a Rate-based Congestion Control Overlay System Table of Contents
A RATE-BASED CONGESTION CONTROL OVERLAY SYSTEM Lianjie Cao, M.S. George Mason University, 2011 Thesis Director: Dr. Brian L. Mark Traditional TCP/IP networks provide a best effort delivery serviceExpand
Improving VPN performance over multiple access links
TLDR
This work describes the systempsilas implementation, the test environment it built on Emulab, and shows that when access links form the communication bottleneck in the end-to-end connection the authors can significantly increase VPN performance over conventional approaches. Expand
Network Integrated Transparent TCP Accelerator
TLDR
The requirements and overheads associated with co-locating middlebox functionality on such computing resources on networking hardware are explored and an example of TCP acceleration proxy (CHART) that improves throughput over networks with delay and loss is used. Expand
Comparison of end-to-end and network-supported fast startup congestion control schemes
  • M. Scharf
  • Computer Science
  • Comput. Networks
  • 2011
TLDR
It is argued that end-to-end fast startup schemes such as an increased initial congestion window and the recently proposed Jump-Start mechanism would be a promising solution for the Internet. Expand
TransCloud - Design Considerations for a High-performance Cloud Architecture Across Multiple Administrative Domains
TLDR
It is argued that a single, flexible architecture, analogous to the TCP/IP stack for networking, is sufficient to support computation jobs run over the wide area and heterogeneous data sets. Expand
...
1
2
...

References

SHOWING 1-10 OF 24 REFERENCES
Major Improvements in TCP Performance Over Satellite and Radio
  • L. Roberts
  • Computer Science
  • MILCOM 2006 - 2006 IEEE Military Communications conference
  • 2006
TLDR
A new signaling message is introduced, exchanged as a flow is setup, which allows the network to specify the highest rate it can support for the flow, thereby allowing TCP to proceed immediately (skipping slow-start) at this rate. Expand
Analysis of link failures in an IP backbone
TLDR
The results indicate that disruption to packet forwarding after link failures depends not only on routing protocol dynamics, but also on the design of routers' architectures and control planes, which is a more appropriate metric for service-level agreements to support emerging applications. Expand
TCP/IP and Linux Protocol Implementation: Systems Code for the Linux Internet
TLDR
This book provides the rapidly growing audience of Linux site managers, as well as researchers and developers worldwide, with the information they need on how Linux TCP/IP keeps the network running. Expand
Quick-Start for TCP and IP
This document specifies an optional Quick-Start mechanism for transport protocols, in cooperation with routers, to determine an allowed sending rate at the start and, at times, in the middle of aExpand
S3: a scalable sensing service for monitoring large networked systems
TLDR
S3, a Scalable Sensing Service, is proposed that achieves the above requirements and enables personalized sensing of the environment as dictated by applications. Expand
A blueprint for introducing disruptive technology into the Internet
TLDR
This paper argues that a new class of geographically distributed network services is emerging, and that the most effective way to design, evaluate, and deploy these services is by using an overlay-based testbed, and suggests four design principles that are not widely supported in existing testbeds. Expand
Exploiting routing redundancy via structured peer-to-peer overlays
TLDR
This paper presents two adaptive mechanisms for structured overlays and illustrates their operation in the context of Tapestry, a fault-resilient overlay from Berkeley, and describes a transparent, protocol-independent traffic redirection mechanism that tunnels legacy application traffic through overlays. Expand
A control-theoretic approach to the design of an explicit rate controller for ABR service
TLDR
A control-theoretic approach to the design of closed-loop rate-based flow control in high-speed networks using a dual proportional-plus-derivative controller that demonstrates the excellent transient and steady-state performance of the controller. Expand
Tapestry: a resilient global-scale overlay for service deployment
TLDR
Experimental results show that Tapestry exhibits stable behavior and performance as an overlay, despite the instability of the underlying network layers, illustrating its utility as a deployment infrastructure. Expand
A Simple Common Sensor Interface for PlanetLab
TLDR
This document came about because of the number of distributed query processors the authors were aware of which were about to be deployed on PlanetLab: at the very least PIER, Sophia, IrisNet, and TAG, and probably Astrolabe are already deployed and in daily use. Expand
...
1
2
3
...