Remember the times when a datacenter IT infrastructure consisted of piles of various hardware, complicated interconnection systems, multiple management tools and software, and an army of system administrators and engineers to maintain and manage all that environment? Well, now, things got simple thanks to the hyperconvergence technology aimed to provide easy management and maintenance for the IT environment. How it does that? With hyperconverged appliances.
But what is a hyperconverged appliance (HCA)?
To put it simply, it’s a box with storage, compute and networking components united inside. It’s a hardware product covered with a single management interface to administrate all the components inside it.
However, the real magic lies in the software-defined everything approach where the software-defined storage (SDS) stack eliminates the need for the proprietary storage hardware, significantly reducing costs. The storage itself is controlled at the OS or hypervisor layer with virtual storage controllers. These virtual controllers run on every node within the cluster ensuring the unified storage management, better resiliency, and failover capabilities. Further on, the software-defined networking (SDN) makes network management agile and flexible by delivering a centralized interface to manage traffic and quickly distribute network resources. It dramatically improves IT infrastructure efficiency by automating the provisioning and configuration of the entire networking stack.
You can find out more about the software-defined storage here: http://www.datacenterknowledge.com/archives/2014/03/05/software-defined-storage-really-mean
So, what do we have? A single building block with all the components united inside of it with SDS serving to deliver useful features and provide a single management interface for the entire solution. Why it’s called a building block? Basically, you get the entire IT infrastructure in a single HCA, or hyperconverged node if you like it more this way, but if you are expanding the infrastructure, you’ll have two, three, or dozens of such nodes. It should also be noted that various vendors provide different integration capabilities for their hyperconverged solutions. For instance, a hyperconverged node may be compatible only with hardware or software of the manufacturer due to its software backend peculiarities. Something that is called a “vendor lock-in”. However, there are solutions that are both hardware- and software-agnostic and simple to integrate into any infrastructure. If you need an example, check out this hyperconverged solution:
Where can I use HCA?
For ROBO locations!
First of all, when you’re opening a new branch or remote office. Why? Well, imagine you’re going with traditional infrastructure where all the components (storage, networking, compute, and management, of course) are separated. You have to investigate the hardware and software for compatibility, buy separate hardware pieces and then configure all that to work as a single system. Takes some time, wouldn’t you agree? On the opposite, you have a hyperconverged appliance already good to start working. You just need to wait for the box to arrive, then you plug it in, and…that’s, actually, it. With HCA, it’s quite simple to open a new office since the IT infrastructure is available from the day one.
For VDI infrastructures!
Hyperconverged nodes are a great choice for running virtual desktop infrastructures (VDI). First of all, HCA scales out by adding more nodes, simultaneously increasing both compute and storage resources. As a result, provisioning new virtual desktops becomes easier as new CPU, RAM, and storage resources are automatically available to the common resource pool. Further, hyperconverged systems usually feature data optimization technologies (say “hello” to software-defined storage) like deduplication and compression. As a result, storage density increases since virtual desktops occupy less space.
For sizing your IT infrastructure!
Hyperconverged appliances are great when it comes to increasing compute power or storage capacity as they provide flexible scaling options. You can scale up by adding more CPUs, disks, or memory till all the slots are full. On the other hand, if you need both capacity and performance to be increased, you simply add nodes, connect them to the network, and your applications get all the necessary resources for proper operation.
The core philosophy around hyperconverged appliances is simplicity. In all its meanings. HCA is simple to manage as it is software-centric and gives a unified and single management tool to work with all the IT environment components. It’s a single box with everything you need to run your applications inside. As a result – less hardware to maintain means less operational expenses. And, finally, it is simple with an HCA to deploy an IT infrastructure from scratch or scale the existing one.