Edge computing speeds up multi-view detection for faster results
Distributed Edge Inference: an Experimental Study on Multiview Detection
Distributed, Parallel, and Cluster Computing
Summary
As internet services grow bigger and more complex, relying on central cloud servers can slow things down. The authors studied a way to split the work across many nearby devices, called edge computing, to make multi-view detection faster. They tested this method with a special software and found it could almost double the speed compared to just using a central server. This means devices can work together better, making some computer tasks quicker and more efficient.
What this means in practice
- •For network engineers: Optimize distributed edge deployments to improve multi-view detection speed under varying network conditions.
- •For embedded systems developers: Implement edge inference frameworks on low-power devices to achieve faster object detection across multiple camera views.
Authors
Gianluca Mittone, Giulio Malenza, Marco Aldinucci, Robert Birke
Abstract
Computing is evolving rapidly to cater to the increasing demand for sophisticated services, and Cloud computing lays a solid foundation for flexible on-demand provisioning. However, as the size of applications grows, the centralised client-server approach used by Cloud computing increasingly limits the applications' scalability. To achieve ultra-scalability, cloud/edge/fog computing converges into the compute continuum, completely decentralising the infrastructure to encompass universal, pervasive resources. The compute continuum makes devising applications benefitting from this complex environment a challenging research problem. We put the opportunities the compute continuum offers to the test through a real-world multi-view detection model (MvDet) implemented with the FastFL C/C++ high-performance edge inference framework. Computational performance is discussed considering many experimental scenarios, encompassing different edge computational capabilities and network bandwidths. We obtain up to 1.92x speedup in inference time over a centralised solution using the same devices.