Jay Livens and Karl Rautenstrach continue their conversation on Azure for cloud
Giovanni Tropeano: Karl you mentioned the cloud first; you know why not cloud attitude shift right the businesses have taken. Companies that are using — you know they start with cloud, then they go to hybrid cloud and now you have businesses are adopting multi cloud even right? So they’re using multiple different clouds for multiple different reasons. How does Azure Stack enable the hybrid cloud for businesses today?
Karl Rautenstrauch: Yeah great question. So Azure Stack really becomes an extension of Azure Public Cloud Resources and I will give you a couple of great examples of where our customers are applying it. And these are only examples they’re not the only ways you can use Azure Stack. We have a company that sends ships out to the sea for an extended periods of time who have heavily invested in Azure Public Cloud Platform. But very, very difficult to reach the public cloud in a usable fashion when you’re out on open ocean relying on satellite connectivity, right it just doesn’t work. So for them what they were able to do is deploy copies of their applications that have been built for Azure Public on Azure Stack, those Azure Stack instances, hardware instances live on the ships and can operate in a disconnected state while they are out and away from port. So it means for them they don’t have to develop two different applications for two different platforms right? It’s, develop once, deploy wherever they need to, Azure Public or Azure Stack those ships.
I have another customer I have worked with who has SLAs and privacy and data security requirements to their customers, where they are actually contractually obligated not to allow that data out of their own data centers. So this is the space where their customers have not yet opened up to the idea of public cloud and haven’t evaluated the security and regulatory requirements. So, they are legally bound to their customers to keep this data on premises. Tremendous benefits for them, leveraging those on-demand resources, we have talked about, but the data has to stay in their data centers. So, a means for them to deploy and get the best of both worlds is to use Azure Stack for some local processing, all data residency in locality, and have Azure Public Cloud, compute resources, access that data in their data centers, over what we call an express route connection, which is a high speed, low latency, dedicated connection from their data center or an exchange provider to Azure Public. So they get the best of all the worlds. High performance burstable compute capabilities from Azure Public Cloud, they don’t have to purchase that compute and put it in their own data center and yet the data never leaves their site. So a couple of examples of how Azure Stack can complement an Azure Public deployment for customers no matter what their use cases may be.
Jay Livens: The other interesting challenge of course is that when you have Azure Stack and you are generating unique data, you obviously have a desire to potentially make copies of that data for development purposes, or even push that data back to leverage the power of the cloud environment, because obviously Azure Stack being a local environment doesn’t have the same extreme scalability that you know true Azure and the cloud does. So, finding ways, using technology to take these copies or take the data and transparently move it to the cloud for further computing or analytics or whatever, I think is an important aspect of value add for the solution, so you get the compute you want when you need it, like guys in the ship, but once the ship is back in port, they probably want to send back some critical information that maybe they want to do some other large scale analytics, or some other computing around it. And so sort of combine the best of Azure Stack on premise with the extreme power and flexibility of Azure Cloud, is really an important element as well.
Karl Rautenstrauch: Absolutely you nailed it. Efficient transport, to and from, is incredibly important. And yeah I mean if you are going to do long-term historical analytics, you want that data as close to the processing as possible. So local processing on the ships in their case, to serve the needs of their customers and then historical funding and advanced analytics in the cloud, perfect example and exactly what they are doing.
Jay Livens: Right.
Learn more about non-traditional ways businesses are leveraging Azure for cloud
Webinar – Protecting Workloads in Azure
Sign up for blog updates via email.Subscribe