In the Software-Defined Data Centre all infrastructure must be programmable so that it can be consumed without leaving the cloud management ecosystem. It should be possible to configure, protect and consume all virtualised and bare metal network, compute and storage resources without leaving System Center. It should be simple to integrate hardware workflows with business workflows. Hitachi Unified Compute Platform (UCP) for Microsoft Private Cloud is a converged infrastructure built with the IT administrator in mind. It features Hitachi Data Systems servers, storage, and network integrated into a pre-validated, pre-configured platform. UCP Director software aggregates and automates the management of physical resources and exposes these for consumption through SCVMM, PowerShell or RESTful API calls. This dramatically simplifies the path to on-demand infrastructure and IT as a Service. This session will take a look at the way that UCP Director integrates seamlessly into the System Center Ecosystem and walk through some typical workflow to show how to seamlessly integrate hardware, hypervisor and application layers.
Also you will have chance to take a look on our validated Scale-Out File Server Solution for Hyper-V and SQL Server leveraging RDMA/RoCE Features on our UCP Stack.
In this talk, we will describe a cloud-centric data science process to go from raw data on premise or in the cloud to operationalized machine learning models using only a browser and elastic resources in Azure using various tools like SQL Server on Azure, Azure VMs, Azure Machine Learning, and Azure HDInsight (Hadoop).
В докладе рассмотрены различные подходы к обеспечению высокой доступности приложения как в публичном, так и в частном облаках. Затронуты темы резервного копирования и построения катастрофоустойчивых конфигураций.