Serial experiments online

Abstract

Current network protocols must comply with rigid interfaces and rules of behavior to fit into well defined, vertical protocol stacks. It is difficult for network designers to offer a wide spectrum of alternative protocols suitable for diverse situations, and to make the stack evolve to match new needs. The tendency is to design protocols that can adapt to the widest possible spread of use. However, even the best adaptive protocols cannot possibly cope with all situations. When their adaptivity limits are reached, the ability to switch to other protocols becomes a clear advantage.Our aim in this paper is to present Lightweight Autonomous resIlient Networks (LAIN), a framework that exploits the multiplicity of alternative protocol, and exposes the spectrum of choice to the advantage of the applications. The system runs continuous experiments with alternative protocols online, in parallel as well as serially, in order to select automatically those that best match the application's needs under the current network conditions. We report first results on the feasibility of the approach and point out the need for such a system in network and protocol evolution.

Evolutionary Autonomous Networks

“…Sentient beings overcome similar problems through online experimentation and continuous learning [55]. Similar approaches have discovered the best networking protocol to deploy in the network [56], the best radio parameters [57], found the best bit-rate for wireless communication [58], or reinvented modern technology [59].…”

Designing Run-Time Environments to Have Predefined Global Dynamics

“…The logic that guides the composition process, however, is not yet well researched. Ramoz-Munoz et al presented an online experimentation environment for network protocols [9], which continuously executes and evaluates the performance of several network protocols, and then selects the best performer. Their approach works well for selecting the best protocol for a specific task or service.…”

Process-in-Network: A Comprehensive Network Processing Approach

“…There are several proposals that seek to make optimal use of the network, by finding the best overlay network topology to minimise the overhead. Within this group are the following solutions: additional auxiliary nodes [ 14 ], hierarchy of nodes [ 15 ], organisational groups [ 16 ], simplification of the network [ 17 ], replication [ 18 ], semantics/users [ 19 ], symbiotic networks [ 20 ], multiple paths [ 21 ], a variety of routes [ 22 ], adaptive/unstructured approach [ 23 ], bandwidth reservation [ 12 ], flexibility in the underlying layer [ 8 ], and traffic prioritisation [ 24 ]; Approaches focusing on information optimisation. This approach attempts to perform some processing of information to make a simplification.…”