A Q&A with Paul Abram, Data Center and Enterprise Compute Practice Leader, and Brett Anderson, Senior Director, HPE Solutions, Logicalis US
Just as “cloud” was the IT industry’s favorite buzz word a few short years ago, today’s tech talk is all about “hyperconverged” solutions – and with good reason. The flexibility and automation built into a hyperconverged infrastructure gives users a cloud-like experience inside their own data centers. But, experts warn, hyperconverged isn’t right for every situation.
Until recently, technology solutions consisted of separate servers, storage and networks configured and physically managed on-premise based on a custom design for every client. Then came converged technology which combined servers and storage according to a reference architecture to deliver a pre-built, pre-packaged solution. The next evolution of converged technology is hyperconverged technology – solutions that integrate servers, storage, networking, virtualization, and automation, orchestration and IT services into a single solution. A key distinction between converged and hyperconverged is the addition of built-in software-defined storage. The question is, which solution is best? To learn more, hear what Logicalis experts Brett Anderson and Paul Abram have to say in this brief Q&A.
Q: Which solution is best – a traditional architecture, a converged infrastructure or a hyperconverged solution?
BA: They are all right for different scenarios. It’s not one or the other – it all depends on what the client is trying to accomplish. Before deciding between them, the client needs to have a vision in mind. They have to know where they are trying to go before they can determine which technological solution will help them get there.
Q: Is all the hype about hyperconverged technology warranted – is this a legitimate solution for customers to consider?
PA: Yes, it is. Hyperconverged technology offers a lot of promise, but you have to align the promise with your specific requirements and the vision of what you’re trying to accomplish. Hyperconvergence is a great solution for specific workloads. If you’re deploying a VDI solution, have requirements for remote branch offices, or need to build a pilot hybrid cloud strategy with automated on-prem-to-cloud workflows, a hyperconverged strategy may be right for you. It’s an exciting set of technologies, but it’s not necessarily for every application workload. We see clients adopting this as part of a multi-year IT strategy, but it is framed to specific use cases and business needs.
Q: What is different about a hyperconverged solution that makes it such a powerful option to consider?
PA: It’s definitely the hyperconverged solution’s flexibility and the inclusion of a software-defined management layer; those are the things that make the software-defined enterprise easy to achieve and cost-effective to scale out. The built-in automation and orchestration capabilities allow customers to take a strong step forward in creating a cloud-like experience within their own data center and to move from a manually intensive second platform operating model into a very dynamic environment that lends itself to third-platform flexibility.
Q: While there are a number of vendors that offer hyperconverged solutions – all strong options in different environments – there is one vendor making particular headlines today with a recent acquisition. Tell me about HPE and SimpliVity and why this is such an important combination.
BA: HPE’s hyperconverged solutions deliver integrated management and IT automation features, while SimpliVity provides policy management, data protection and backup capabilities with file-level restore and guaranteed data efficiency. Their strengths are complementary and will provide a very powerful hyperconverged infrastructure solution when merged.
Q: What kinds of situations should trigger a CIO to consider a hyperconverged solution?
PA: Maybe they’re undergoing a data center refresh because of a merger, acquisition or aging technology, for example, or perhaps they’re refreshing branch office computing environments. The key to determining if a hyperconverged solution is right for their specific situation, however, lies in clearly defining the needs of the business. This means they need to have a multi-year IT roadmap at least framed out before they can determine what kind of solution will work best for their organization.
Q: Where should a CIO start when trying to weigh the pros and cons of a hyperconverged solution?
PA: Start with the application layer – what applications (i.e., SAP) or business services (i.e., email) do they need to run? How business-critical are these applications and what are their performance demands? Growth considerations? Answering these questions will help them determine what their core requirements really are.
They also need to think about the kinds of requirements they have in their environment. Hyperconverged solutions, for example, are designed for virtualized environments, so CIOs should ask themselves if they have hypervisor or physical requirements in their project scope.
Q: People say hyperconverged infrastructures are well suited to expansion – is that true?
BA: Yes, but before deciding on a direction, it’s important to consider your compute, memory and storage requirements to ensure you’ll get the performance you require. How many users do you have today? What are your growth plans? In general, it’s true that one of the top advantages to a hyperconverged infrastructure is the ability to expand linearly as business needs change – and to do it cost effectively.
But there’s more to it than that. While looking at your resources and considering growth paths are important considerations that will extend the life of the solution, it’s also important to know your networking, security and audit requirements – and to match them to the right hyperconverged solution.
You also need consider your applications. Applications don’t operate in isolation; they integrate with other systems in the business environment, which means it’s critical to fully consider the ways these technologies will interact with each other and within the context of your cloud strategy because certain hyperconverged infrastructures may align better with these requirements than others.
Q: What about data protection – does a hyperconverged solution offer any kind of data protection for the user?
PA: Yes, each version of hyperconverged infrastructure does provide some level of data protection. The question is, how does that align to your business’ data protection strategy? You really have to define your backup and recovery service requirements before shopping for a hyperconverged solution to be sure you’re covering all the bases.
Q: What else does an IT pro need to think about before deciding on a hyperconverged infrastructure?
PA: They need to consider their operational and management requirements. Understanding the business’ requirements around alerting, updating and logging are critical steps toward success with hyperconverged solutions.
BA: And migration paths are also an important consideration. Once you’ve implemented a hyperconverged solution, how will you get from here to there? You’ve got to define your migration strategy before moving forward, and determine up front if you’ll need a professional services partner to assist. Logicalis offers these kinds of migration services, and we have a whole professional services team ready to help if an organization wants to outsource their migration from one platform to another.
Want to Learn More? Find out why HPE’s Flexible Capacity is a solid option for hyperconverged solutions in this blog post by Logicalis expert Brett Anderson. Next, examine the six areas that must be addressed in a digital-ready reference architecture, then download a Logicalis white paper, “Converged Infrastructure for Targeted Workloads.” Finally, explore the results of Logicalis’ 2016 CIO survey, “Digital Enablers: The Challenges Facing CIOs in an Age of Digital Transformation”: http://ow.ly/fyz7308hHVH.