Timo Hönig

TL;DR: In this article, a fine-grained energy-overhead analysis of software mitigations of the Meltdown and Spectre vulnerabilities is presented, which reveals application-specific energy overheads of up to 72 %.

TL;DR: In this paper, a wait-free approach is proposed to arbitrate the plentiful concurrent strands managed by a concurrency platform's runtime system by exploiting inherent properties of fully-strict fork/join concurrency.

TL;DR: A modular analysis approach, EnergyBudgets, which bridges external energy measurement hardware to the Linux perf subsystem and shows that energy budgets accurately measure the energy consumed by different workloads and allow for an overhead-reduction on the SUT by 20% to 51% in comparison to regular timers, while still guaranteeing the same level of precision.

TL;DR: Precious as discussed by the authors is an approach, as well as practical implementation, that estimates execution time and power draw of neural networks that execute on a commercially available off-the-shelf accelerator hardware (i.e., Google Coral Edge TPU).