Resource Central, a machine learning framework for Azure

We are working on Resource Central, which is a large-scale machine learning framework for Azure. Our paper describing Resouce Central has just been accepted for publication at SOSP 2017. Here is the final copy:

Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms Eli Cortez (Microsoft); Anand Bonde (Microsoft Research); Alexandre Muzio (ITA, Brazil); Mark Russinovich, Marcus Fontoura (Microsoft); Ricardo Bianchini (Microsoft Research)

As part of this work we also release a public dataset that characterizes the Azure workloads. This repository contains a representative subset of the first-party virtual machine workload (VM) of Microsoft Azure in one of its geographical regions. We hope this data is useful to foster more research in resource management for the cloud.