Predictive VM (Virtual Machine) auto-scaling is a promising technique to optimize cloud applications’ operating costs and performance. Understanding the job arrival rate is crucial for accurately predicting future changes in cloud workloads and proactively provisioning and de-provisioning VMs for hosting the applications. However, developing a model that accurately predicts cloud workload changes is extremely challenging due to the dynamic nature of cloud workloads. Long- Short-Term-Memory (LSTM) models have been developed for cloud workload prediction. Unfortunately, the state-of-the-art LSTM model leverages recurrences to predict, which naturally adds complexity and increases the inference overhead as input sequences grow longer. To develop a cloud workload prediction model with high accuracy and low inference overhead, this work presents a novel time-series forecasting model called WGAN-gp Transformer, inspired by the Transformer network and improved Wasserstein-GANs. The proposed method adopts a Transformer network as a generator and a multi-layer perceptron as a critic. The extensive evaluations with real-world workload traces show WGAN- gp Transformer achieves 5× faster inference time with up to 5.1% higher prediction accuracy against the state-of-the-art. We also apply WGAN-gp Transformer to auto-scaling mechanisms on Google cloud platforms, and the WGAN-gp Transformer-based auto-scaling mechanism outperforms the LSTM-based mechanism by significantly reducing VM over-provisioning and under-provisioning rates.

The protection of cultural heritage and property is a significant and critical task that requires collaboration and expertise in a variety of disciplines. Of the many risk factors, insect infestation is one cause of deterioration and loss. At a large, state university, disparate departments, ranging from Facilities Management to the Entomology Department and Veterinary Medicine, assisted the university museum in identifying a drywood termite infestation, determining the extent of loss and developing a plan to prevent or mitigate future infestations. Our group was able to determine the extent and severity of a drywood termite infestation in the museum storage vault through visual inspection and X-ray computed tomography (CT). This paper describes the process and heuristics of identifying and estimating the amount of active/inactive termite infestations in the art frames as well as visualizing a 3-dimensional structure to learn the extent of infestation. This interdisciplinary collaboration and effectual use of tomography enabled our group to determine the condition of several art frames through non-invasive means and develop a plan of action to identify and prevent future insect incursions within the museum.

No abstract

Predictive Virtual Machine (VM) auto-scaling is a promising technique to optimize cloud applications' operating costs and performance. Understanding the job arrival rate is crucial for accurately predicting future changes in cloud workloads and proactively provisioning and de-provisioning VMs for hosting the applications. However, developing a model that accurately predicts cloud workload changes is extremely challenging due to the dynamic nature of cloud workloads. Long-Short-Term-Memory (LSTM) models have been developed for cloud workload prediction. Unfortunately, the state-ofthe-art LSTM model leverages recurrences to predict, which naturally adds complexity and increases the inference overhead as input sequences grow longer. To develop a cloud workload prediction model with high accuracy and low inference overhead, this work presents a novel time-series forecasting model called WGAN-gp Transformer, inspired by the Transformer network and improved Wasserstein-GANs. The proposed method adopts a Transformer network as a generator and a multi-layer perceptron as a critic. The extensive evaluations with real-world workload traces show WGAN-gp Transformer achieves 5× faster inference time with up to 5.1% higher prediction accuracy against the stateof-the-art approach. We also apply WGAN-gp Transformer to auto-scaling mechanisms on Google cloud platforms, and the WGAN-gp Transformer-based auto-scaling mechanism outperforms the LSTM-based mechanism by significantly reducing VM over-provisioning and under-provisioning rates.