An Efficient Monte Carlo-Based Probabilistic Time-Dependent Routing Calculation Targeting a Server-Side Car Navigation System

Abstract

Incorporating speed probability distribution to the computation of the route planning in car navigation systems guarantees more accurate and precise responses. In this paper, we propose a novel approach for dynamically selecting the number of samples used for the Monte Carlo simulation to solve the Probabilistic Time-Dependent Routing (PTDR) problem, thus improving the computation efficiency. The proposed method is used to determine in a proactive manner the number of simulations to be done to extract the travel-time estimation for each specific request while respecting an error threshold as output quality level. The methodology requires a reduced effort on the application development side. We adopted an aspect-oriented programming language (LARA) together with a flexible dynamic autotuning library (mARGOt) respectively to instrument the code and to take tuning decisions on the number of samples improving the execution efficiency. Experimental results demonstrate that the proposed adaptive approach saves a large fraction of simulations (between 36% and 81%) with respect to a static approach while considering different traffic situations, paths and error requirements. Given the negligible runtime overhead of the proposed approach, it results in an execution-time speedup between 1.5x and 5.1x. This speedup is reflected at infrastructure-level in terms of a reduction of around 36% of the computing resources needed to support the whole navigation pipeline.

Pegasus: Performance Engineering for Software Applications Targeting HPC Systems

“…KAP is responsible for providing K path alternatives for routing a vehicle from origin to destination. PTDR incorporates speed probability distribution to the computation of the route planning in-car navigation systems to guarantee more accurate and precise responses [39]. BC provides information about centrality nodes in the routing map (a graph) needed to identify critical nodes.…”

“…The NavSys code version used in this paper has been developed by the Czech supercomputing center IT4I to provide an experimental testbed for extending the existing Sygic navigation by server-side routing with a traffic flow calculation for global optimization of city transportation. The NavSys application is a result of recent research on its main components, such as path reordering [40], kalternative paths [41,42], betweenness centrality [43,44,45], and probabilistic time-dependent routing [39]. Although the complete NavSys application is not publicly available, two of the important components codes, PTDR [39] and BC [43,44], have been disclosed and are available online 5 .…”

EVEREST: A design environment for extreme-scale big data analytics on heterogeneous platforms

“…(2) the optimization goal set for execution (e.g., performance or energy consumption) [35], [36], (3) the additional dynamic requirements (e.g., security monitoring, data features [37]), and (4) the available techniques for data management (e.g., data representations and distributed allocation). The selection will generalize the concept of affinity between the code variants and the available system configurations and requirements.…”

Dynamic Network selection for the Object Detection task: why it matters and what we (didn't) achieve

“…We then try to create a predictor function that can work as an oracle for unknown images. This is a proactive approach that relies on the concept of input feature that is present in some works in literature [24,13]. To create the predictor, we will search for some data features and we use them to create a function that can predict which is the best network to use to perform the inference.…”