What is HCI?
Many organisations today are running traditional 3 tier architectures consisting of compute, storage and networking. While this architecture has served organisations well, advances in software means it’s no longer the best approach. It is expensive to build, complex to operate, difficult to scale and can’t respond quickly enough to meet today’s application demands.
In order to gain more agility businesses often consider public cloud services, however, concerns about losing control, higher costs and security mean that often this isn’t a viable option. Enter Hyperconverged Infrastructure (HCI). Modern hardware with higher performance memory and CPU’s combined with advanced software-based storage technology let organisations converge their entire infrastructure and run it all on X86 based servers – this is HCI.
HCI is a software-defined IT infrastructure that virtualizes all of the elements of conventional hardware-defined systems. At a minimum it includes virtualized computing, a virtualized SAN and virtualized networking. It’s a scale-out platform composed of shared software-defined storage and compute running on commodity x86 Hardware, managed with a common software management plane.
Getting to HCI
Virtualization has been the default deployment in the data centre for many years, however, how that is being deployed has changed significantly in that timeframe. Moving on from 3 tier architecture, a converged infrastructure improves on this by bringing compute, storage, management and networking into a single rack. Hyperconverged solutions take this to the next step by converging these infrastructure silos into industry standard servers and virtualizing physical infrastructure. So HCI enables a competitive business through the software-defined cloud enabled datacenter by natively integrating the functions running them all on software on a virtualized platform along with management tools.
The key benefits of HCI over traditional 3 tier architecture includes:
Operational simplicity: HCI reduces operational costs with an integrated hardware and software solution. The consolidation of administration functions into a simple unified management framework allows compute and storage resources to be monitored and managed in less time freeing up personnel to focus on other priorities.
Lower Total Cost of Ownership (TCO): The use of scale up/scale out architecture with industry standard X86 servers means that hyperconverged systems have a lower cost of entry and total cost of ownership compared to legacy systems. Organisations can start with the minimum requirements and grow the infrastructure non-disruptively in a granular fashion.
Greater agility: HCI delivers day 0 support for the latest technologies providing the agility to increase application performance and rapid application deployment for business-critical applications without major infrastructure change.
Automation: HCI also supports infrastructure automation as HCI workflows involve fewer steps that traditional storage workflows and are designed to be automated and integrated from the outset, not as an afterthought.
Use cases for HCI
There are a number of reasons why organisations opt for a hyperconverged infrastructure solution but, ultimately, it boils down to the specific needs and demands of the business. Many choose HCI for the ability to manage the system remotely – which is particularly convenient if there are multiple offices scattered across various locations. Typical use cases include:
VDI: Initially suggested as the ideal use case for HCI, while this is still true and typically a perfect use case many server-based workloads make good candidates for running on HCI Infrastructure. HCI is starting to become the default design choice for a large proportion of new deployments. The question has shifted to “why can’t I run this workload on HCI?”
Private Cloud: HCI delivers the onsite (private) cloud with the benefits of public cloud including flexibility and accessibility so is a good option to rapidly create a private cloud and maintaining the ability for IT to monitor and manage the network and still provide easy access.
Pilot/Proof of concept: When new business requirements come up, HCI can be used to scale applications. The advantage of using HCI boxes is that they integrate very easily with cloud infrastructure and can very quickly extend your storage to accommodate new requirements. In the same way HCI can also simplify the IT management of an organisation’s remote or branch locations.
HCI may not be the holy grail for your datacentre modernisation and there are some key areas you should consider before going down this route.
HCI may not be the ideal platform for some environments. For example, this could be an environment where the amount of storage required is not expected to grow linearly as compute does, or, if a requirement of one of the four key metrics (compute, memory, storage capacity or storage performance) is significantly different to the other capacity requirements.
Organisations also need to be mindful of the general compute capacity available per host may be lower than previously expected as some resource is assigned on each host to run the storage component of HCI. This may demand purchasing higher specification CPU’s or accepting a less dense virtual machine to host ratio.
Organisations with a mature and well understood storage infrastructure may need to re-evaluate HCI during the next server and storage refresh cycle as changing to it when a large investment has already been made may not make financial sense.
The main players
Currently, there are three main players when it comes to HCI – VMware, Nutanix and DellEMC who together have over 50% of the market share and are the named leaders in Gartner’s Magic Quadrant for Hyperconverged Infrastructure 2018.
It’s clear that HCI offers a number of benefits to the modern datacenter. One of the compelling qualities of hyperconvergence is that it’s easy to start small, migrate a little bit at a time and grow as needed until the migration is complete, taking away the painful and disruptive process of infrastructure removal and replace. Organisations that are due for a hardware refresh or embarking on a greenfield deployment should investigate if HCI is a suitable replacement for the traditional three tier stack.
Before deciding on which type of infrastructure to implement, organisations should however first analyse exactly the nature of the workloads they wish to run on it. Knowing the exact needs of the business will go a long way to ensuring a smooth transition and efficient outcome with minimal wasted investment. For many this can be a long and daunting process but working with a services partner such as Xtravirt can help you plan and map out the transition to a hyperconverged infrastructure and ensure that you get a solution that is right for your organisations needs.
Xtravirt has been involved in transforming IT infrastructure and hyperconverged solutions for many years and brings a broad range of skills and experience to the table. If you’re looking to transform your datacentre, we can provide advisory, design and implementation services to create the right solution for your organisation. Contact us and we’ll be happy to assist you.
About the Author
James Kilby joined the Xtravirt consulting team in November 2018. He has over 10 years of hands-on experience in identifying and engineering business solutions across a number of technology stacks including VMware and Microsoft based technologies. James is an active member of the VMware User Group community and a founding member of the UK Veeam User Group. He is a VMware vExpert and Veeam Vanguard.