October 02, 2017

Hyperconverged in a Cloud World

by: Tarun Bhasin, Research Director

Datacentre Infrastructure, IDC Canada

 

Hyperconverged in a cloud world-min.jpg

In a previous blog, my colleague David Senf talked about the importance of cloud and automation in this era of digital transformation (DX). While cloud and hybrid cloud are topics that have been widely discussed over the past few years, it is worth spending some time looking at the rapid rise of a new infrastructure model that many IT users have been adopting. This new infrastructure model — known as hyperconverged Infrastructure (HCI) systems — is one of the fastest-growing segments in the infrastructure market, with many of the large public and private cloud environments today being built on these systems.

 

For IDC, HCI systems fall within our broader converged systems market coverage. HCI is one of the four product categories that we cover within the converged segment. HCI systems are differentiated from the other converged systems because HCI systems natively collapse core storage, compute, and storage networking functions into a single software solution or appliance. Thus, one key characteristic of hyperconverged systems that differentiates these solutions from other integrated systems is their ability to provide all compute and storage functions through the same server-based resource. This is in contrast to other converged systems, which are essentially a combination of separate compute, storage, and networking systems at the factory by the vendor or by resellers into a single form factor. All HCI systems also employ a distributed file system or an object store that serves as the data organization, management, and access platform, and a hypervisor that hosts the hardware abstraction layer and provides workload adjacency, management, and containerization.

 

What Workloads Does HCI Support?

This rapid adoption of HCI means that worldwide revenue for HCI in 2016 was over $2.3 billion, more than double the 2015 revenue. Revenues in Canada were up by 115% in 2016. And while some of the early adoption for HCI was driven by the IT infra-centric workloads like virtual desktop, HCI systems today are deployed for a wide variety of workloads. A recent survey of 500 organizations across Canada reveals that 3 of the top 5 HCI workloads today are Big Data & analytics, application & software development and collaboration & end-user productivity workloads.

 

Will HCI Adoption Impact How My IT Department Operates?

Since HCI systems are different from the traditional server storage and networking deployments, their adoption means some changes are required to the organizational structure and roles. The consolidation of hardware often leads to a breakdown of siloes that are built around the traditional IT infrastructure resources, as well as changes in roles and accountabilities. Traditionally siloed roles such as server or storage administrators will be consolidated with a broader scope of responsibilities. For example, a systems administrator may take on a broader infrastructure administrator role that covers compute, storage, and virtualization. There may be additional requirements for a "cloud architect" role. IT staff will need to shift their focus from managing the physical resources like servers and storage arrays to managing applications and services, the cloud environment, and virtual machines.

 

Also, since many of the new applications are developed for the cloud, and the old tools used to manage the infrastructure (for HA, DR replications, etc.) are essentially run separately for the cloud and the physical infrastructure, the new HCI environment will enable organizations to standardize on a single set of tools that will work irrespective of the infrastructure setup. The fact that all the systems are in a single rack also provides more visibility and enables the use of fewer tools. All this means fewer tasks and less time spent on managing the resources.

 

What Should I Think About Before I Implement HCI in My Organization?

For organizations looking to implement HCI solutions, here are some key considerations:

  • HCI may not be an optimal fit for all your workloads. Consider starting small so that you have an opportunity to understand the impact on your organizational structure and processes before you move to a broader deployment. It is important to identify a workload that makes sense so that the users and stakeholders can see the benefits and are on board for the broader deployment. Based on IDC research, the top workloads that organizations are looking to move to HCI include unstructured content/data analytics, app development and testing, and VDI.
  • Do a rigorous evaluation process of vendors before choosing your supplier. This is an emerging and fast-growing segment, there are a number of vendors playing in this space, and not all the solutions out there will work for your specific requirements. Consider vendors that specialize in the converged systems space rather than vendors that provide solutions across the datacentre. Also look for vendors that provide flexibility beyond the hardware. This includes flexibility in pricing, support/service levels, and even software-only deployments.
  • IDC research indicates that support and service are key buying criteria when choosing a converged vendor. Make sure you give enough consideration to prospective vendors' track record on service and support.
  • Understand the needs of both the IT and the business users. From an IT perspective, understand the server, storage, and networking needs, as well as the growth and transition plans for the workloads over the next few years. Take into account the cost involved around licensing and maintenance of the HCI infrastructure. From a business perspective, spend time to better understand the long-term strategy and requirements of the business users so that you can map the business strategy to the infrastructure environment over the medium to longer term

Compare NetApp HCI vs. Other Hyper Converged Systems

 

 

 

About the Author

Tarun.jpg

Tarun Bhasin is Research Director, Datacentre Infrastructure on the Infrastructure & Cloud Solutions team. In his role, he provides market share, market sizing and segmentation analysis and identifies and analyzes market trends, vendor strategies, customer behavior, and technologies affecting the server market in Canada.  In addition, Tarun is also the regional lead for the Quarterly Server Tracker, the Quarterly Converged Systems Tracker and the Quarterly Cloud IT Infrastructure Tracker. IDC quarterly trackers are the premier tool for analyzing, measuring and forecasting various product markets, with the most accurate and timely information on changes, trends, and activities impacting the market.