Resource Provisioning Using A Clearing House Architecture

C-N. Chuah, L. Subramanian , R. H. Katz and A. D. Joseph

[Poster Slides in PDF] [IWQoS'2000 Paper]


Introduction

There has been increasing research effort in making the Internet architecture capable of providing different service levels to meet QoS requirements. Integrated Services (Int-Serv) with RSVP signaling introduces end-to-end per-flow reservations, which requires core routers to maintain individual flow states and may not scale well. Differentiated Services (Diff-Serv), on the other hand, relies on packet markers and per-hop behavior (PHB) at core routers to provide service differentiation. Diff-Serv bandwidth brokers (DSBBs) are proposed to negotiate resource reservations between neighboring domains. However, admission control is only performed at the ingress edge router to a Diff-Serv domain, and reservations are set up for aggregate traffic on hop-by-hop basis without any global knowledge of the network conditions. Diff-Serv trades off end-to-end QoS for scalability.

The lack of a well-studied policy architecture to regulate resource provisioning in a scalable manner has motivated our design of a Clearing House (CH) as an alternative solution. The CH attempts to provide better QoS assurance and higher network utilization, as offered by stateful networks (e.g., Int-Serv), while maintaining the scalability of a stateless network architecture (e.g., Diff-Serv). We also address two other problems that are often encountered today: (a) lack of a methodology for intra- and inter-domain provisioning or service composition based on dynamic traffic patterns, and (b) lack of distributed resource controlling systems for large domains. Diff-Serv bandwidth brokers regulate inter-domain reservations using a flat architecture with one broker per domain, regardless of the size and population of the domains. Our initial simulation results show that such architecture does not scale for large domains, and causes a single point of congestion.

The Clearing House Architecture

Our CH architecture uses a hybrid of a flat and a hierarchical structure. A hierarchical structure is deployed within large domains to help distribute the network state information among the various CH-nodes and reduces the amount of states maintained. On the other hand, a flat structure is used for peer-to-peer provisioning across multiple domains. At the top level, our architecture appears flat while the hierarchical structure is associated with large ISPs/ASs, or a subset of adjacent domains that share mutual trust.

We have also developed a distributed controller for large domains that attempts to maximize the effective throughput seen by the entire system and adapts to fluctuating load patterns. The CH-nodes close to the host networks are responsible for performing admission control. The edge routers maintain only aggregated state information about the flows and the core routers are completely stateless. The CH-nodes keep track of the intra- and inter-domain traffic patterns, and adapt aggregate reservations dynamically based on "Gaussian traffic predictors". The CH architecture can inter-operate with MPLS, OSPF and other queuing mechanisms like Core-Stateless Fair Queuing (CSFQ). The CH can be used to provision an ISP for VPN or VoIP traffic, and to achieve better QoS assurance across multiple domains.

Initial simulation results show that a single-node CH employing FIFO scheduling for the vBNS backbone topology has a saturation point of 1800 reservation requests per second for traffic load with Poisson arrival. Using aggregate scheduling, the throughput increases to 3500 reservation requests per second and the mean response time reduces by a factor of 4. Our simulations based on actual audio traces show that dynamic reservations based on a one-minute Gaussian predictor can achieve average packet loss of less than 1%, while incurring only 7% over-provisioning.

The Roles of the Clearing House in ICEBERG

The CH-architecture plays two different roles in the ICEBERG project:

A. Third-Party Clearing House
First, CH can be implemented as a third-party distributed architecture that manages the traffic flow between the ICEBERG Network Plane and the ISP Plane. The CH can provide secure billing services to ICEBERG, and coordinates resource provisioning across multiple domains (ISPs/ASs) to provide statistical QoS assurance. The CH-nodes closer to the host networks are responsible for:

One of the major design requirements is to make CH scale over both distance and number of ISPs to support the wide-area operation of ICEBERG. Two key ideas behind the CH design are hierarchy and aggregation. Assuming that ICEBERG is operating over a subset of IPSs/ASs that share mutual trust and agreements among one another, we can implement the CH as a hierarchical distributed system. In our model, physically adjacent basic domains are aggregated to form bigger logical domains, and a CH-node is associated with each domain. This introduces a hierarchy of logical domains and an associated CH-tree. At each level, multiple CH-nodes share the tasks of (a) monitoring incoming traffic and network performance, (b) storing reservation status, and (c) perform aggregate reservations within the domains. For wide-area calls that cross multiple domains, a recursive call will be made to the parent-CH nodes at the upper CH-tree to adapt inter-domain aggregate reservations. To preserve scalability, reservations are only performed for aggregate flows. Only the local CH performs per-flow admission control and provides secure billing to individual calls. Parent CH-nodes only maintain aggregate billing statements. Each domain aggregates its billing tickets and sends them to the CH for lump settlement payments at the end of regular billing cycle.

B. Local Clearing House for Load Balancing
A local CH-hierarchy can be deployed within ICEBERG to choose the optimal path to achieve load balancing among different operators, e.g., speech recognizer, text-to-speech converter, or gateways, e.g., H.323 gateways.

[Home]