What is a next-generation hyperconverged solution and why consider it?

March 6, 2018 Sam Chapman

It’s clear that many companies have benefited from deploying Hyperconverged infrastructure solutions. Compared to Direct Attached Storage (DAS), Networked Attached Storage (NAS), and Storage Area Networks (SAN) solutions that have been traditionally deployed over the past 20 years, HCI delivers clear benefits such as simplified deployment, improved performance, and ease of management. However, with recent innovations in Hyperconverged technology, there is a distinction between 1st and next generation HCI solutions.

Let’s consider…

1st-generation HCI solutions give you the flexibility to combine compute and storage on a single node, deploy a combination of compute only nodes, flash, or hybrid HDD & Flash solutions that address the needs of your applications. However, over the past five years companies have experienced challenges with their deployments of 1st-generation HCI solutions due to their limited ability to scale, performance limitations, and limits on resiliency. A vast majority of these deployments have been limited to the small/medium business sector due to these constraints and many customers have needed to deploy these HCI solutions within silos in order to manage the requirements of their applications. Let’s take a look at some of the limitations of a 1st-generation HCI solution.

Storage Efficiency and Resilience

1st-generation HCI solutions that position themselves as primary or secondary storage typically mirror data across nodes in the cluster. When deployed, the administrator has to decide on a replication factor in order to protect the data that is being copied across the nodes. Typically, 2x or 3x replication in order to support multiple drive failures or loss of a node. The dis-advantage of this strategy is that you give up storage efficiency and performance in order to ensure resiliency. At 3x replication factor, over 50% of the raw disk capacity is utilized across the cluster to handle all the data being copied. Imagine deploying an all flash system where 50% of the capacity is being leveraged for copies of the data? This would be an expensive proposition that negates many of the cost benefits of HCI, impacts performance, and limits the ability to scale the cluster.

Managing Multiple Workloads

Traditional HCI solutions typically have a challenge in managing multiple workloads on a single cluster. Storage administrators are forced to still size their deployments based on peak workload performance requirements. They aggregate the performance requirements, and “oversize” the solution to ensure they don’t run out of resources…or even worse, they begin to create “application silos” to separate workloads and ensure there is no contention. When this is done, many times excess CPU, Memory, and Storage capacity is never fully utilized. This is another challenge that negates the benefits of 1st generation HCI solutions.

Scalability and Performance

1st-gen HCI has always touted the promise of scalability by being able to grow the size of your HCI cluster one node at a time, as your requirements grow. As you add nodes you can either increase compute power or capacity depending on your application requirements. The key challenge for many storage administrators, is that as you increase the size of the cluster, the performance of the overall cluster increases…however, so does the overhead of mirroring data via the assigned replication factor. There is also a very limited ability to apply resources to your mission critical applications when needed. Another major limitation of 1st generation HCI solutions. More power plus more overhead and less control = less benefit.

What if?

What if… there was an HCI solution that leveraged patented erasure coding Technology and provided you with the resiliency of mirroring data across nodes, however could provide you with up to 94% efficiency as you scaled the size of the cluster? Up to 94% Usable Capacity with minimal management overhead. What if you could lose up to 3 drives and an entire node and the system would continue to provide maximum performance? The benefit would be less equipment, data center space, power, & cooling costs.

What if… you were provided with a Quality of Service Engine that enabled you to align performance requirements with your business needs? And your mission critical applications were assigned policies that guaranteed minimum IOPs, Throughput, and latency? And you could schedule the availability of system resources for Mission Critical Applications during peak hours? The benefit would be optimal application performance, better user experience, and removing the need to “over-size” your HCI Cluster.

What if… there was an HCI solution that truly leveraged next generation NVME PCIe Technology from the ground up and could scale with next generation 3D X-Point and Intel Optane Technology? And also leveraged I/O Data Reduction algorithms to extend the life of your SSDs by 4x, reduce capacity requirements in half, and reduce latency by 50%? What if there was an HCI system that leveraged a QOS Engine that Dynamically Placed Data on the fastest storage available based on utilization of that data and defined policies? The benefit would be improved reliability, less hardware, and a dramatically improved user experience.

What if… you had the ability to take practically unlimited Snapshots and Replicate into the Cloud for business continuance and disaster recovery? Or creation of Clones that can be mounted as VMs to conduct test and development without impacting your production environment? The benefits here would be improved reliability, flexibility, and improved quality.

Now, what if you could deploy this system very easily, Define Policies with 2 clicks, and manage the system via a simple vCenter Plug-In or a CLI and be able to customize the system via industry standard APIs? The benefits would be less time managing your server and storage infrastructure…and more time providing Business Value.

What if… this were actually available right now?

That’s right. The next generation of Hyperconvergence is already here…with Pivot3. If you’re asking the question: How can I leverage the benefits of Pivot3 Technology, I already have an existing 1st generation HCI solution deployed?” Reach out to your local Pivot3 Solution Architect and they can walk you through the process of deploying a new solution or leveraging your existing HCI hardware. If you would like to learn about how Pivot3 has deployed over 500 PBs of storage, on over 18,000 Appliances, at over 2,500 customers in 54 countries, visit the Pivot3 HCI Zone, reach out to a Pivot3 Partner, or contact your local Pivot3 Sales/Technical Professional.


See Pivot3’s Next Generation HCI Solution live for Video Surveillance & Data Center Applications!

The post What is a next-generation hyperconverged solution and why consider it? appeared first on Pivot3.

No Previous Articles

Next Article
Challenging What’s Possible with HCI
Challenging What’s Possible with HCI

Pivot3 recognized as a Challenger in the first-ever Gartner Magic Quadrant for Hyperconverged Infrastructur...