Alan Stanley John Chapman

Aggregate flow control: improving assurances for differentiated services network

The differentiated services architecture is a simple, but novel, approach for providing service differentiation in an IP network. However, there are various issues to be addressed before any sophisticated end-to-end services can be offered. This work proposes an aggregate flow control (AFC) technique with a Diffserv traffic conditioner to improve the bandwidth and delay assurance of differentiated services. A prototype has been developed to study the end-to-end behavior of customer aggregates. In particular, this new approach improves performance in the following manner: (1) fairness issues among aggregated customer traffic with different number of micro-flows in an aggregate, interaction of non-responsive traffic (UDP) and responsive traffic (TCP), and the effect of different packet sizes in aggregates; (2) improved transactions per second for short TCP flows; and (3) reduced inter-packet delay variation for streaming UDP traffic. Experiments are also performed in a topology with multiple congestion points to show an improved treatment of conformant aggregates, and the ability of AFC to handle multiple aggregates and differing target rates.

Enhancing TCP performance with a load-adaptive RED mechanism

This paper describes a technique for enhancing the effectiveness of RED by dynamically changing the threshold settings as the number of connections land system loadr changes. Using this technique, routers and switches can effectively control packet losses and TCP timeouts while maintaining high link utilization. Copyright

A scalable and distributed WWW proxy system

With the wide-spread popularity of the World-Wide Web (WWW), network and server load increases dramatically. Caching proxies have been introduced to improve the system performance with the assumption that a page will be fetched many times before it is destroyed or modified. However the increase in user requests has driven up to the limit of the proxy performance. The authors present a distributed proxy server architecture that can increase the service availability, provide system scaleability, coupled with load balancing capability. The system employs TCP-based switching mechanism which has a finer session granularity, and more dynamic control on resource allocation.

A Connectionless Approach to Providing QoS in IP Networks

The attempt to provide QoS in IP networks has raised some interesting questions on how a service can be provided to meet the application requirements while obeying the network resource constraints. Previous efforts focussed on a flow-based, connection oriented approach to deliver QoS for IP Networks — Intserv. This approach was quite comprehensive but it has not been widely deployed because of complexity and scalability issues. A recent packet marking based scheme called Differentiated Services (Diffserv) Architecture provides a relatively simple and coarse approach. It is too early to predict the usefulness of this approach. This paper outlines a framework to deliver IP QoS which is based on Intserv. It addresses scalability concerns by removing the need for a connection-oriented reservation setup mechanism and replaces it with a Diffserv-like mechanism to consistently allocate bandwidth end-to-end in a network. A prototype device is discussed that manages bandwidth on a node. An algorithm is presented that allows the device to automatically detect application QoS requirements without the need for application-level signalling. A priority-based scheduling mechanism with a variant of weighted round-robin is described.

Experimental study of assured services in a diffserv IP QoS network

Much attention has recently been given to the differentiated services (Diffserv) approach to provide Quality of Service (QoS) for IP networks. This packet-marking based approach to IP QoS is attractive due to its simplicity and ability to scale. Two of the most popular services proposed for the Diffserv approach are the Assured and Premium Services. In this work prototypical implementations of Diffserv components are described. The prototypes are used to study the single-queue, dual drop-preference model proposed as a basis for assured services in Diffserv.

A load adaptive mechanism for buffer management

Abstract The basic idea behind active queue management schemes such as random early detection (RED) is to detect incipient congestion early and to convey congestion notification to the end-systems, allowing them to reduce their transmission rates before queues in the network overflow and packets are dropped. The basic RED scheme (and its newer variants) maintains an average of the queue length which it uses together with a number of queue thresholds to detect congestion. RED schemes drop incoming packets in a random probabilistic manner where the probability is a function of recent buffer fill history. The objective is to provide a more equitable distribution of packet loss, avoid the synchronization of flows, and at the same time improve the utilization of the network. The setting of the queue thresholds in RED schemes is problematic because the required buffer size for good sharing among TCP connections is dependent on the number of TCP connections using the buffer. This paper describes a technique for enhancing the effectiveness of RED schemes by dynamically changing the threshold settings as the number of connections (and system load) changes. Using this technique, routers and switches can effectively control packet losses and TCP timeouts while maintaining high link utilization.