Success with Hybrid Cloud: Defining “Hybrid”

Success with Hybrid Cloud: Defining “Hybrid”

  • Comments 6
  • Likes

Overview

In any region, at any size, in any industry, an enterprise’s IT infrastructure has an enormous impact on the success of that organization. There are as many infrastructure setups and strategies as there are enterprises, with each IT team tailoring it to the unique and changing needs of that business. At Microsoft we are realistic about how quickly and dramatically the needs of a datacenter can change, and that is why we enable a Hybrid Cloud IT model.

This post is the first in a multi-part series that this blog will feature as a detailed overview of Microsoft’s Hybrid Cloud approach to enterprise IT. This series will examine the building/deployment/operation of Hybrid Clouds, how they are used in various industries, how they manage and deliver different workloads, and the technical details of their operation.

The reason I think a Hybrid Cloud is the best strategy for any organization is simple: It can accommodate the way your organization grows, organizes, and operates – in any setting, under any circumstances.

To start, I want to get specific about what Microsoft means when we use the term “Hybrid Cloud” by sharing a few concrete examples:

First, one of the most common hybrid scenarios we are seeing is the use of Windows Azure for a tier of an application. Increasingly often, organizations are moving the web tier of an application to Azure and leaving the middle and data tiers in their datacenter. This setup allows the web tier to seamlessly scale on-demand while the data is kept within the customer’s firewalls. This powerful and scalable combination of public and private cloud resources is the Hybrid Cloud model.

Another example can be seen in how organizations are approaching their disaster recovery needs. Many service providers are now offering business continuity solutions leveraging Hyper-V Replica to back up data from their customer’s datacenter to the service provider – or as a geo-replication between a service provider’s datacenter.  Customers are also using Azure as a backup location (this is a topic I’ve discussed previously – here and here).

Getting the maximum value for this type of scenario is easy, thanks to the fact that Windows Server 2012 and 2012 R2 include a license to use Windows Azure Backup (which provides 5GB free backup for each server). In a DR scenario, the Hybrid Cloud gives a customer a complete backup in the cloud that is always available – and this comes at a prince point that is far more reasonable than the traditional tape-based solution.

The Microsoft Hybrid Cloud DR/storage solution got even better last November when we acquired StorSimple. With StorSimple, you can now backup all or selected portions of your data to the cloud. This is something we refer to as cloud-integrated storage – and if you haven’t tried it already, I really recommend it. Cloud-integrated storage allows you to set policies that can take some really specific actions, like leaving the hot blocks of data local and moving the cold blocks to Azure. Then, if a request comes in for the data from the cold blocks it is rapidly served up from Azure. This effectively gives you a bottomless datacenter because you have simple access to an effectively unlimited amount of cloud storage.

These are just a couple of examples of the Hybrid Cloud model, and, over the next several weeks, I will cover many more in greater depth. I’m also going to cover the technical elements that need to be set up and configured to enable these scenarios. For example, how and why you have to seamlessly stretch your network across clouds, as well as how we’ve made BYOIP (Bring Your Own IP Address) simple in Windows Server 2012 R2 so that you can cloud deploy servers that come configured with hard-coded IP addresses.

In this introductory post, let me answer one important question: What factors are driving organizations toward adopting Hybrid Cloud solutions?

First, this approach helps organizations avoid placing all of their eggs in one basket. Having all of your data in a single place makes you vulnerable to the occasional outage, and it also puts you at the mercy of your access to a particular physical or virtual location.

A critical example of this was Hurricane Sandy back in the Fall of 2012. While this storm was wreaking havoc on the east coast, we received feedback from multiple customers who noted that even though their datacenters were literally under water, their businesses kept running without a hitch because they had used Hyper-V Replica. All of their VMs were replicated to an external cloud. This is a great example of a Hybrid Cloud, and this is one of the many reasons that most of the organizations I meet with say they will be (or will continue to) use multiple clouds in the future.

Second, many organizations operate in a highly-regulated industry where the regulatory constraints may dictate where and how they store data. A purely public cloud approach is not likely an option here whereas a Hybrid Cloud may offer the best solution for keeping certain data local while still relying on the flexibility and scale of the public cloud to manage demand.

Third, public clouds offer incredible agility, and agility is the entire reason to use the cloud in the first place. Cloud computing also offers a bargain pricing for infrastructure that is globally accessible, always available, and (in Microsoft’s case) has an iron-clad, monetary-based SLA supporting the availability of your applications. I say “in Microsoft’s case” because Amazon and others will not offer an SLA backed by monetary penalties if their up-time SLA is missed.

Fourth, a lot of companies are asking how much of the technical burden of the cloud should be put on their shoulders – especially for businesses who don’t have (or want!) to include the assembly and maintenance of datacenter infrastructure as a core competency. The Hybrid Cloud model removes these technical roadblocks: We have invested over $15 billion in the last three years building out a global network of datacenters that give any organization the worldwide reach and access to any level of technical resources – whether they need to dive deep into their technical resources, or deploy a plug-and-play PaaS solution. These are core competencies for Microsoft.

The growth and pattern of innovation makes Microsoft’s Hybrid Cloud a very unique thing in the technology industry. We are the only organization in the world that is operating a global, at-scale public cloud and then applying everything we learn into the products we ship. These products are in turn used by enterprises and service providers who then build their own clouds. Putting this technology in the hands of our customers is a very different thing than charging a monthly subscription for a static service.  This flexibility and choice is the real power of the Cloud OS.

A core part of this approach is ensuring the Microsoft clouds deliver consistency and seamless mobility so that you are never locked into any single cloud.

Another unique element of Microsoft’s approach is that we are the only public cloud provider operating 200+ at-scale SaaS offerings on that public cloud infrastructure. The reason this is important is simple: Applications drive infrastructure improvements. Windows Server, for example, is a better operating system because of what we learn every day from Azure and these 200+ services running on Windows Server. This means that the applications you get from us to use in your datacenters are also better because the same teams that build and operate these cloud services also deliver applications like Exchange, SharePoint, SQL and System Center (to name a few).

This is all part of our “cloud-first” design principle. We architect and build every single one of our products with cloud scale, security, availability, and operability in mind. We are passionate about innovating and creating new value and new solutions, and we test these features exhaustively within our own cloud services. When these features have been battle-hardened and proven effective, we then deliver them to you in our on-premises products. For an enterprise or a service provider looking for a platform that is stable, reliable, and world-class – this is a big deal.

This approach is not only cloud first, but it is also customer first

Looking ahead to the coming weeks of this series, you can see an overview of every future “Success with Hybrid Cloud” blog post here.

Your comment has been posted.   Close
Thank you, your comment requires moderation so it may take a while to appear.   Close
Leave a Comment
  • As data is important so cloud.....will make it cloudyy