Chemical algorithms are statistical control algorithms described and represented as chemical reaction networks. They are analytically tractable, they reinforce a deterministic state-to-dynamics relation, they have configurable stability properties, and they are directly implemented in state space using a high-level visual representation. These properties make them attractive solutions for traffic shaping and generally the control of dynamics in computer networks. In this paper, we present a framework for deploying chemical algorithms on field programmable gate arrays. Besides substantial computational acceleration, we introduce a low-overhead approach for hardware-level programmability and re-configurability of these algorithms at runtime, and without service interruption. We believe that this is a promising approach for expanding the control-plane programmability of software defined networks (SDN), to enable programmable network dynamics. To this end, the simple high-level abstractions of chemical algorithms offer an ideal northbound interface to the hardware, aligned with other programming primitives of SDN (e.g., flow rules).

In wireless sensor and actor networks (WSANs), nodes collaborate to accomplish distributed sensing and computation. One of the most attractive features is that global macro- behaviors in the network emerge from local micro-actions in and interactions between nodes. In this paper, we review a recently proposed approach, which bases on a chemical metaphor and its related theory, to design, analyze, and implement systems with such an emerging property. As an example of distributed, emerging computation in WSNs, we extend an existing gossip-like chemistry-inspired protocol (the “Disperser”) and propose a new chemically driven solution to consensus problems that is applicable in WSNs. We analyze it with a signal processing approach and validate results through simulations. The proposed chemical algorithm is not based on the iterative packet-exchange but rather on a continuous rate-modulation. The algorithm is thought to lie at the physical layer and to directly control the basic transmission hardware of the nodes in a sort of protocol-less operation mode.

The Content Centric Network (CCNx) protocol introduces a new routing and forwarding paradigm for the waist of the future Internet architecture. Below CCN, a volatile set of transport and flow-control strategies is envisioned, which can match better different service requirements, than current TCP/IP technology does. Although a broad range of possibilities has not been explored yet, there are already several proponents of strategies that come close to TCP's well-known flow control mechanism (timeout-driven, window-based, and AIMD- operated). In this paper, we carry out an empirical exploration of some proposed strategies in order to assert their feasibility and efficiency. Our contributions are twofold: First, we establish if receiver-based, timeout-driven, AIMD operated flow-control on Interest transmissions is sufficiently effective for CCN in a future Internet deployment, where it may co-exist with TCP. In this process we compare the performance of three different variants of this strategy, in presence of multi-homed content in the network (one of them proposed by the authors). Second, we provide indicators for the general efficiency of timeout-based flow-control at the CCN receiver, in presence of in-network caching, and exhibit some of the challenges faced by such strategies.

Traffic-shaping algorithms, such as congestion avoidance or rate limiting for cloud services, control the dynamics of packet flows in a distributed setting. However, despite their significance in networking, it is still hard to design them, to predict their dynamic behavior, and to prove their robustness. With the help of a simple chemical metaphor, we recently proposed a novel engineering framework for designing, executing, and analyzing traffic-shaping algorithms: by letting packets react akin to chemical molecules, the design of (distributed) flow-control algorithms becomes drawing reaction networks. In this way, we gain in analyzability: the related fluid model, describing the emergent behavior of the overall system, can be derived from the corresponding reaction networks automatically. In this article, we describe how to fine-tune chemical algorithms. We therefore analyze their transient behavior in order to reveal the semantics and sensitivity of key parameters. Methods from different domains assist us in this endeavor: We first linearize the fluid model as proposed in Metabolic Control Analysis, we then describe it in control-theoretic terms, and finally, we characterize its sensitivity in the frequency domain. We demonstrate the feasibility and reveal the limits of our method by applying it to a chemical algorithm for distributed rate control.

A challenging problem when engineering network and communication protocols is to predict and formally describe the dynamic behavior of a protocol. Chemical Networking Protocols (CNPs) offer the opportunity to transpose the good and well-understood analyzability of chemical reactions to networking protocols, not only at the design stage of protocols, but also for the study of their dynamic behavior. In this paper we focus on this latter aspect and we demonstrate how signal processing techniques from classical signal and control theory can be employed in the study of the average flow properties of CNPs. Our analysis methods include a model linearization as proposed in Metabolic Control Analysis, a statespace description classically used in control theory, and the system's characterization in the frequency domain, which is central to signal theory. We demonstrate the feasibility of our method by applying it to an actual congestion control protocol that models TCP's behavior. With our contribution, CNP designers can easily determine the key parameters of their protocols and understand how to calibrate them in order to obtain the desired behavior.