Hyper-convergence is the current and potentially future course of a journey that began years ago with server and storage virtualisation to optimise the traditional siloed approach to information technology.
Hyper-convergence is basically an emerging technology in which the server, storage and networking equipment are all combined into one software-driven appliance. Hyper-converged infrastructure (HCI) can be managed comfortably through an easy-to-use software interface. Typically, a hyper-converged platform uses standard off-the-shelf servers and storage, and a virtualisation hypervisor that helps abstract and manage these resources conveniently. A simple operational structure makes this infrastructure flexible, easy to scale and manage.
HCI is gaining traction not just in enterprises but also with others who can benefit from heavy computing without complex management, such as professional racers! Training and racing is usually data-intensive. Racers need information like speed, acceleration and torque in real time to understand how they are performing and how they need to proceed. One mistake could cost not just a victory but a life too! Racing speed depends on the speed and reliability of the data processing and analytics that takes place behind the scenes. Last year, Formula One racing team Red Bull Racing replaced its legacy systems with HPE Simplivity hyper-converged infrastructure, achieving 4.5 times faster performance, increased agility and lower TCO through the move.
A few quick facts about HCI
At the outset, hyper-convergence might sound similar to all the virtualisation stuff you have read about in the past. So, let us first set out some facts about HCI before rounding up the latest updates:
1. Nutanix, a leader in HCI, defines that the technology streamlines deployment, management and scaling of data centre resources by combining x86-based server and storage resources with intelligent software in a turnkey software-defined solution.
2. HCI combines compute, storage and network together as a single appliance. It can be scaled out or expanded using additional nodes. That is, instead of scaling up the traditional way by adding more drives, memory or CPUs to the base system, you scale out by adding more nodes. Groups of such nodes are known as clusters.
3. Each node runs a hypervisor like Nutanix Acropolis, VMware ESXi, Microsoft Hyper-V or Kernel-based Virtual Machine (KVM), and the HCI control features run as a separate virtual machine on every node, forming a fully distributed fabric that can scale resources with addition of new nodes.
4. Since most modern HCI are entirely software-defined, there is no dependency on proprietary hardware.
5. HCI does not need a separate team as it can be managed by any IT professional. This makes it ideal for small and medium enterprises.
6. HCI is different from the traditional server and shared-storage architectures. It also differs from the previous generation of converged infrastructure.
7. HCI is not the same as cloud computing!
8. Scale-out and shared-core are two of the keywords you are bound to encounter when reading about HCI. Basically, most HCI implementations involve multiple servers or nodes in a cluster, with the storage resources distributed across the nodes. This provides resilience against any component or node failure. Plus, by placing storage at the nodes, the data is closer to compute than a traditional storage area network. So, you can actually get the most out of faster storage technologies like non-volatile memory express (NVMe) or non-volatile dual in-line memory module (NVDIMM).
The scale-out architecture also enables a company to add nodes as and when required rather than investing in everything upfront. Just add a node and connect it to the network, and you are ready to go. The resources are automatically rebalanced and ready to use.
A lot of hyper-converged implementations also have a shared core, that is, the storage and virtual machines compete for the same processors and memory. This reduces wastage and optimises resource usage. However, some experts feel that in some special cases, users might have to buy more equipment to run the same workloads.
9. It’s true that HCI is great for small enterprises as it can be used without fuss and hassles, but it can be used by really large data centres too. Leading companies have on record hyper-convergence case studies where more than a thousand HCI nodes are installed in a single data centre.
10. In a recent survey, research leader Forrester found that the most common workloads being run on hyper-converged systems are: database, such as Oracle or SQL server (cited by 50 per cent); file and print services (40 per cent); collaboration, such as Exchange or SharePoint (38 per cent); virtual desktop (34 per cent); commercial packaged software such as SAP and Oracle (33 per cent); analytics (25 per cent); and Web-facing workloads such as LAMP stack or web servers (17 per cent).