Providing service guarantees requires strict admission control to ensure that the resources are not over-allocated. On another hand, limiting the core nodes' role has been recognized as a key aspect in providing scalable quality of service (QoS) solutions. This paper presents a scalable multiple-token mechanism that ensures a distributed and coordinated resource allocation along the logical ring of the ingress/edge nodes. The proposed mechanism alleviates the scalability problem by keeping all the admission control operations in the ingress/edge nodes while preserving guaranteed service semantics. By distributing the available bandwidth of the links over multiple tokens, increased robustness and low latency under light load conditions are achieved, while the zero-false positives property is maintained. The erroneous connection denials, which could be due to the distribution of available bandwidth information, are avoided by an accumulation procedure that concentrates the available bandwidth information whenever necessary.

Providing service guarantees requires strict admission control to ensure that the resources are not over-allocated. On another hand, limiting the core nodes' role has been recognized as a key aspect in providing scalable quality of service (QoS) solutions. This paper presents a scalable multiple-token mechanism that ensures a distributed and coordinated resource allocation along the logical ring of the ingress/edge nodes. The proposed mechanism alleviates the scalability problem by keeping all the admission control operations in the ingress/edge nodes while preserving guaranteed service semantics. By distributing the available bandwidth of the links over multiple tokens, increased robustness and low latency under light load conditions are achieved, while the zero-false positives property is maintained. The erroneous connection denials, which could be due to the distribution of available bandwidth information, are avoided by an accumulation procedure that concentrates the available bandwidth information whenever necessary.

A method for determining the highest available-bandwidth paths in a core-stateless network, with or without restrictions, is proposed. The method keeps track of the possible solution range as it progresses, can operate under number of hops, delay or cost constraints, and induces load balancing and high network resource utilization. The method is combined with a token-based transfer of information and semaphore mechanism to avoid over-allocation and ensure resource allocation coordination along the logical ring of the ingress nodes. The resulted connection admission control procedure alleviates the scalability problem by keeping all the admission control operations in the ingress nodes while preserving the guaranteed service semantics and providing a low-latency, quasi-parallel allocation under light load and a centralized/semaphored allocation under heavy load. The widest path search algorithm is illustrated and compared with other existing algorithms.

Assuring guaranteed services and providing fair bandwidth sharing are much sought characteristics in the current network architecture research. The IntServ approaches achieve those characteristics but only at the expense of using per-flow states and signaling, which are not scalable and difficult to implement. Alternatively, the stateless approaches proposed up to now are focused either on providing guaranteed services similar to those offered by IntServ or on ensuring a reasonably fair bandwidth allocation. In this paper we propose a stateless architecture - the fair bandwidth allocation control (Fair-BAC) - that ensures both the adherence to the given throughput guarantees and a fair allocation of the remaining bandwidth between the flows that do not have any reservation and the out-of-profile fraction of the flows with reservation. The capabilities of Fair-BAC are demonstrated in a set of simulations performed for various degrees of congestion and using sources with different conformance levels.