Driven by technologies such as IoT-enabled health care, machine learning applications at the edge, and industrial automation, mobile edge and fog computing paradigms have reinforced a general trend toward decentralized computing, where any network node can route traffic, compute tasks, and store data, possibly at the same time. In many such computing environments, there is a need to cache significant amounts of data, which may include large data sets, machine learning models, or executable code. In this work, we propose a framework for joint computation scheduling, caching, and request forwarding within such decentralized computing environments. We first characterize the stability region of a “genie-aided” computing network where data required by computation are instantly accessible, and develop a throughput optimal control policy for this model. Based on this, we develop a practically implementable distributed and adaptive algorithm, and show that it exhibits superior performance in terms of average task completion time, when compared to several baseline policies.

We introduce the problem of optimal congestion control in cache networks, whereby both rate allocations and content placements are optimized jointly. We formulate this as a maximization problem with non-convex constraints, and propose solving this problem via (a) a Lagrangian barrier algorithm and (b) a convex relaxation. We prove different optimality guarantees for each of these two algorithms; our proofs exploit the fact that the non-convex constraints of our problem involve DR-submodular functions.

The emergence of IoT devices and the predicted increase in the number of data-driven and delay-sensitive applications highlight the importance of dispersed computing platforms (e.g. edge computing and fog computing) that can intelligently manage in-network computation and data placement. In this paper, we propose the DECO (Data-cEntric COmputation) framework for joint computation, caching, and request forwarding in data-centric computing networks. DECO utilizes a virtual control plane which operates on the demand rates for computation and data, and an actual plane which handles computation requests, data requests, data objects and computation results in the physical network. We present a throughput optimal policy within the virtual plane, and use it as a basis for adaptive and distributed computation, caching, and request forwarding in the actual plane. We demonstrate the superior performance of the DECO policy in terms of request satisfaction delay as compared with several baseline policies, through extensive numerical simulations over multiple network topologies.

Complex systems such as smart cities and smart power grids rely heavily on their interdependent components. The failure of a component in one network may lead to the failure of the supported component in another network. Components which support a large number of interdependent components may be more vulnerable to attacks and failures. In this paper, we study the robustness of two interdependent networks under node failures. By modeling each network using a random geometric graph (RGG), we study conditions for the percolation of two interdependent RGGs after in-homogeneous node failures. We derive analytical bounds on the interdependent degree thresholds (k1,k2), such that the interdependent RGGs percolate after removing nodes in Githat support more than kjnodes in Gj(∀i, j ∈ {1, 2}, i ≠ j). We verify the bounds using numerical simulation, and show that there is a tradeoff between k1and k2for maintaining percolation after the failures.