Asynchronous distributed parallelization of mobile network optimization algorithms

Abstract

It has been years since the introduction of the Dynamic Network Optimization (DNO) concept, yet the DNO development is still at its infant stage, largely due to a lack of breakthrough in minimizing the lengthy optimization runtime. Our previous work, a distributed parallel solution, has achieved a significant speed gain. To cater for the increased optimization complexity pressed by the uptake of smartphones and tablets, however, this paper examines the potential areas for further improvement and presents a novel asynchronous distributed parallel design that minimizes the inter-process communications. The new approach is implemented and applied to real-life projects whose results demonstrate an augmented acceleration of 7.5 times on a 16-core distributed system compared to 6.1 of our previous solution. Moreover, there is no degradation in the optimization outcome. This is a solid sprint towards the realization of DNO.

Acceleration and visualization of Dynamic Network Optimization

“…As an NP-hard multiple-objective constrained combinatorial optimization problem, MNO problems exhibit high complexity that requires applying iterative heuristic algorithms. Our prior work has managed to accelerate two core MNO algorithms, Intelligent MNO Algorithm [6] that is an iterative heuristic algorithm with intelligent learning capabilities and greedy MNO algorithm [7] that only begins upon the completion of the "Intelligent" algorithm. We proposed and implemented novel predictive binary or greedy tree patterns to expose the parallelism.…”

“…Our long-term goal is to realize DNO on a sub-hourly basis, implying that the optimization phase of a DNO cycle ought to complete ideally within a quarter of an hour regardless of the network size and complexity. Our prior work [6] [7] has designed and implemented scalable distributed parallel solutions for two core MNO algorithms that achieved substantial speed gain at 7.5 and 14.5 on a 16-core distributed system. GEOson (originally known as ariesoACP) from Arieso Ltd [8], the market-leading automated MNO product, serves as the testbed for acceleration in our work.…”