The soaring use of machine learning leads to increasing processing demands. As data volume keeps growing, providing classification services with good machine learning performance, high throughput, low latency, and minimal equipment overheads becomes a challenge. Offloading machine learning tasks to network switches can be a scalable solution to this problem, providing high throughput and low latency. However, network devices are resource constrained, and lack support for machine learning functionality. In this paper, we introduce IIsy -a novel mapping tool of machine learning classification models to off-the-shelf switches. Using an efficient encoding algorithm, enables fitting a range of classification models on switches, co-existing with standard switch functionality. To overcome resource constraints, adopts a hybrid approach for ensemble models, running a small model on a switch and a large model on the backend. The evaluation shows that achieves near-optimal classification results, within minimum resource overheads, and while reducing the load on the backend by 70% for data-intensive use cases.

Code offloading enables resource-constrained devices to leverage idle computing power of remote resources. In addition to performance gains, offloading helps to reduce energy consumption of mobile devices, which is a key challenge in pervasive computing research and industry. In today's distributed computing systems, the decision whether to execute a task locally or remotely for minimal energy usage is non-trivial. Uncertainty about the task complexity and the result data size require a careful offloading decision. In this paper, we present Voltaire- a novel scheduler for sophisticated energy-aware code offloading decisions. Voltaire applies machine learning methods on crowd-sourced data about past executions to accurately predict the complexity and the result data size of an upcoming task. Combining these predictions with device-specific energy profiles and context knowledge allows Voltaire to estimate the energy consumption on the mobile device. Thus, Voltaire makes well-informed offloading decisions and carefully selects local or remote execution based on the expected energy consumption. We integrate Voltaire into the Tasklet distributed computing system and perform extensive experiments in a real-world testbed. Our results with three real-world applications show that Voltaire reduces the energy usage of task executions by 12.5% compared to a baseline scheduler.