CINECA case study: the supercomputers that are pushing HPC boundaries
High-performance computing at CINECA is enabling scientists to process and transform huge amounts of data, bringing insightful results...
The term ‘hyper-convergence’ is a growing trend in the IT industry. So, what exactly does it mean? And what are the benefits it can bring to your business?
Hyper-convergence is undoubtedly a hot topic in the IT world right now so here’s how it works – and what it can do for your businesses.
Industry analyst company Gartner believes hyper-convergence is destined for huge growth in 2016. The market for hyper-converged integrated systems (as Gartner calls it) will grow 79 per cent to reach nearly $2 billion (£1.4 billion / €1.8 billion) in 2016, and will become mainstream within the next five years.
Hyper-convergence is, according to Gartner, a “platform offering shared compute and storage resources, based on software-defined storage, software-defined compute, commodity hardware and a unified management interface”.
Lenovo’s Tikiri Wanduragala, EMEA x86 Server Systems Senior Consultant, recently explained to Think Progress how he and his team defines it, and how it differs from a more traditional converged infrastructure.
“[With converged infrastructure] you could choose the storage fabric, server types and networking switches – whatever suited your needs,” he said. “Hyper-converged offers these locked down: fixed storage, fixed networking switches, fixed server capabilities and so on. Then it virtualises the whole thing. So what you now have is like a building block, which is excellent for deploying things quickly.
“The big thing about hyper-convergence is the linking and clustering of the storage units. Until now, storage boxes sat separately from the server and the two talked to each other through clever protocols. In the hyper-converged world, the storage sits inside the server, making use of the intelligence that’s there.”
As it sits so closely to software-defined, the benefits a hyper-converged infrastructure can bring are similar. Ultimately it provides an infrastructure that is much more flexible and easier to manage.
This focus on software and control/management means that’s where the intelligence lies, so hyper-converged infrastructures can be run on commodity x86 hardware, dramatically reducing costs for customers. It also enables businesses to scale up and down as needed.
What it’s being used for
Hyper-convergence is still very much in its early days of adoption and, as a concept, it didn’t really exist until a few years ago. But that hasn’t stopped some early adopters from realising the benefits it can bring and therefore beginning their hyper-converged journey.
One of those is Scottish law firm MacRoberts. The company employs 200 people in its offices in Edinburgh, Glasgow and Dundee. Its IT infrastructure is spread across the Edinburgh and Glasgow offices, but its previous systems were overburdened and operating beyond capacity, the company told Computer Weekly. In addition, management was difficult and scaling when needed proved a challenge.
MacRoberts selected hyper-converged infrastructure from Gridstore, and installed one appliance at each of its sites. It now runs 75 VMs across its Edinburgh and Glasgow sites, which both have a storage capacity of 112TB. What MacRoberts has got out of its hyper-converged infrastructure is a much more simplified management platform, increased resiliency and the ability to scale up when the need arises. It has also saved physical space and reduced its electricity bills thanks to fewer appliances in its data centres.
How to start your own journey
If your business has decided that hyper-converged is the way to go, ensure the project is fully planned before you begin. As Jan Willem de Nooij, Client Technical Specialist at Lenovo, has previously explained on the company’s blog, there’s a four-step process that can help with a hyper-converged project.
Maintaining lots of different software and systems goes against the very idea of hyper-converged and software-defined, so it’s important to minimise the amount you use and stick to a few that will work with the applications you’ll be running on your hyper-converged infrastructure.
This means much more flexibility and scalability with workloads.
Tasks such as deploying virtual machines or adding applications are much more efficient if there is no IT involvement, which reduces operating costs.
The pay-for-what-you-use model can reduce costs, and the cloud can add to the flexibility and manageability that a hyper-converged infrastructure provides.
Because hyper-converged as a concept is still relatively young, it may not be right for all businesses at the moment. However, there’s no doubt that with the right implementation, it can bring big benefits, helping to create an infrastructure that’s cheaper to run, easier to manage and more flexible.
As the market for it develops and matures over the next three to five years, IT departments should familiarise themselves with hyper-convergence and whether it would be beneficial for their companies.