Adaptive load sharing for network processors

Abstract

A novel scheme for processing packets in a router is presented, which provides for load sharing among multiple network processors distributed within the router. It is complemented by a feedback control mechanism designed to prevent processor overload. Incoming traffic is scheduled to multiple processors based on a deterministic mapping. The mapping formula is derived from the robust hash routing (also known as the highest random weight-HRW) scheme, introduced in K.W. Ross, IEEE Network, 11(6), 1997, and D.G. Thaler et al., IEEE Trans. Networking, 6(1), 1998. No state information on individual flow mapping needs to be stored, but for each packet, a mapping function is computed over an identifier vector, a predefined set of fields in the packet. An adaptive extension to the HRW scheme is provided in order to cope with biased traffic patterns. We prove that our adaptation possesses the minimal disruption property with respect to the mapping and exploit that property in order to minimize the probability of flow reordering. Simulation results indicate that the scheme achieves significant improvements in processor utilization. A higher number of router interfaces can thus be supported with the same amount of processing power.

Locating Performance Monitoring Mobile Agents in Scalable Active Networks

“…However, there are some limitations for using mobile agents in Active Networks. One of these restrictions is limited processing power of active nodes ( [10]- [12]). As the result, active nodes have limited power for hosting mobile agents and performing their management tasks.…”

A cluster-based active router architecture supporting video/audio stream transcoding service

“…Although a number of adaptive load sharing policies are proposed in the literature [12], we found that the extended HRW technique [6] is suitable for network applications containing a number of flows. Hence, we implemented it in our system.…”

“…As a result, most of the implementations are based on simple round-robin schemes. In this paper, we identify and implement a particular adaptive load balancing scheme [6] for our active router. We also propose a new stream-based round-robin policy that is simple but greatly reduces the out-of-order degree of the media units in a simple unit-based round-robin scheme.…”

“…It is reported to provide good performance and low overhead when the request identifier space is evenly distributed. Lukas Kencl et al HRW is extended with a feedback mechanism, which allows adjustment to the weights with minimum flow re-mapping [6], to cope with request identifier space locality. We incorporate this adaptive load balancing technique in our active router.…”

Scheduling real-time multimedia tasks in network processors

“…It was shown that round robin is simple and fast, but provides no guarantee to the playback quality of output streams because it causes out-oforder departure of processed media units [7]. Adaptive load sharing scheme, which we implemented from the literature [8], achieves better unit order in output streams, but involves higher overhead to map the media unit to an appropriate node. The aim of this paper is to derive a theoretical framework for load scheduling in a multimedia active router built with a network processor that comprises of a number of processors for packet processing.…”