The paper "On network-aware clustering of web clients" presented at SIGCOMM ’00 by Krishnamurthy and Wang proposes clustering techniques to move content closer to the clients which are generating most of the traffic.

Before the paper, cluster of clients were simply /24 networks. The authors propose to group clients by BGP prefixes.

The paper is focused on web traffic and to determine the quality of the BGP clustering, the authors analyses web server logs. Based on the logs, they map client IP addresses to BGP prefixes.

The paper "An Overview of Routing Optimization for Internet Traffic Engineering" presented by Wang et al. in IEEE Communications Surveys volume 10, N°1 presents a complete survey of the current techniques for traffic engineering.

In this blog, we will not summary the survey, just present the structure.

The paper divides TE according to four criteria.

1. scope: interdomain vs intradomain
2. enforcement: MPLS-based vs IP-based
3. timescale: online vs offline
4. type: unicast vs multicast

The paper "A Measurement-Based Analysis of Multihoming" proposed by Akella et al. at SIGCOMM'03 shows different aspects of performances improvements in multihomed environments.

The journal article we discussed in [08-03-17] is basically the same as this paper, so that we only present a fast summary of the paper.

The paper "Pathload: a measurement tool for end-to-end available bandwidth" presented at the PAM'02 conference by M. Jain and C. Dovrolis presents an tool to estimate the available end-to-end bandwidth.

The Pathload tools is build on the Self-Loading Periodic Streams (SLoPS) (see [08-03-17]).

The paper "On-the-fly TCP path selection algorithm in access link load balancing" written by Lin et al. proposes a new way to implement load balancer based on NAT.

Traditional load-balancer uses weighted round robin to define the uplink to use for a given stream. The model proposed by the authors deals only with TCP flows.

The paper "Practical large-scale latency estimation" written by Szymaniak et al., presents an architecture able the estimate latency of millions of nodes in an efficient manner.

The idea of the paper is to propose CDNs like Google to determine the best server for any client.

To do so, the authors uses GNP, a landmarks-based network coordinate system. First, they establish a base of 7 redundant landmarks which computes their coordinates based on active probing.

The paper titled "IPv6 delay and loss performance evolution" written by Zhou et al. in the International Journal of Communication Systems proposes to analyze the performances of IPv6 paths compared to IPv4 paths between the same nodes.

The dataset is the RIPE database between the October 1st, 2003 and October 31th 2005. The principle is simple: process the RIPE database for the TTM boxes running both IPv4 and IPv6 and compare the performances in term of delay and loss ratio for the two protocols.