Protect your mission critical systems with Hyper-Convergence
28 Oct 2021, by Slade Baylis
They say that nothing in life is guaranteed, except for death and taxes. Here at Micron21 our aim has always been to provide the best cloud services possible, and so we want to add reliability of your services with us to that list. In line with that, we are proud to have achieved the highest possible certification available for data centre reliability and redundancy, making us Australia’s first Tier IV data centre. However, that top-tier reliability for power and cooling won’t matter much if your server has crashed and your applications are offline. This is why it’s becoming more and more common for businesses to look to protect their systems through some form of “High Availability”.
What is High Availability?
High Availability (HA) is any sort of service that allows you to have redundancy of either the server itself, or the application that’s being hosted. Though in other contexts the word “redundancy” usually has a negative connotation, when it comes to IT infrastructure, it’s exactly the opposite. Redundancy of IT systems means that there are backup systems in place, so that if failures occur, the overall system can continue to function as per usual. An example of a HA service would be if you had two identical servers running side by side, with identical versions of the application running on each, as well as an automatic fall-over in place should one fail. Then, if one fails, the other picks up the slack and your application/website continues to function as per normal.
However, the problem with the above scenario is that it’s usually quite expensive. Setting up such an environment requires purchasing at least twice as much infrastructure as you actually need. The reason for this is that you need to have redundancy in each and every system that’s in use, so that if it fails, the secondary system can pick up the slack. So right out of the gates, this means that you are usually looking at double the cost of a system that doesn’t have this form of protection built in! This can be a huge hurdle for smaller businesses to overcome, and often hard to make the financial case for it unless it’s 100% required.
The good news though, is that we’ve been able to incorporate this same type of HA protection into our regular VMware Cloud Server (VCS) platform. Not only are all of our VCS services provided as a HA service by default, but we’ve also kept the costs down and affordable when compared with other solutions out there. This allows for even small businesses to be able to be protected against hardware caused outages, without the sky-high costs that are usually required. How did we do that you may ask? Well, it’s through a technology called Hyper-Converged Infrastructure (HCI).
What is Hyper-Converged Infrastructure (HCI)?
With regular Virtual Private Servers (VPS), whilst they are virtual and in some ways abstracted from the physical hardware (i.e the host on which they run), they still rely on the underlying hardware being in working order. If the physical server has a hardware fault, such as the CPU failing, then the virtual machines running on it will go offline, as will all the services they host. This reliance on the underlying physical server means that even if you have extra redundancy for infrastructure, you still have a single point of failure with the host itself that could take your services offline.
There are multiple different approaches to mitigating that risk, but all of them share a common strategy. That is, making sure that all systems have redundancy built into their design, so that secondary systems can be switched to in the event of failure. For our HA services, we utilise technology provided by VMware for our high-end enterprise-grade services. VMware is the market leader in virtualisation technology and is trusted by over 500,000 customers globally, with 100% of the Fortune 500 companies running on VMware infrastructure. The specific technology that we use to provide that high-availability protection is called Hyper-Convergence.
Hyper-Converged systems avoid the problem of servers being a single point of failure by enabling storage, computing, and networking from multiple physical servers to be combined into a single system. What this provides for is even greater levels of abstraction and automation than a regular VPS service. With this system, any single hardware failure (including total loss of the physical host) will result in the virtual machines starting in seconds on another server. This provides a layer of high availability as opposed to regular VPS platforms. It also enables server maintenance works to occur without downtime, which is a common pitfall for alternative systems.
Whilst the protections against downtime are definitely the highlight when it comes to HCI, another thing to keep in mind is how this technology also enables scalability. The consolidation of servers into clusters means that businesses are able to scale-up their infrastructure as needed, without needing to worry about the underlying hardware required.
This complete abstraction from the hardware that the servers run on, makes it a true cloud service in every meaning of the word. Not only that, but the use of HCI, enterprise-grade Dell hardware, and our Tier IV data centre infrastructure make our VMware services some of the most reliable on the planet.
High-end protection without the usual high-end costs
With the technologies used and the scale of infrastructure needed for this type of protection, you would be forgiven for pre-emptively dismissing it as too expensive. Usually these sort of systems would cost a fortune if you were to set them up yourself. Not only do you need up-front capital investment to cover three top-end servers at a bare minimum, but you would also need to pay for the licencing costs for the software. That’s not even counting the expertise required to set it all up. Knowing this, it made sense for us to go the extra effort and build our VMware Cloud Server (VCS) services to include this as a standard feature so that all our customers could benefit from it.
Our opinion is that too often the word “cloud” gets thrown around without care. Sometimes it refers to highly resilient systems, but more often than not, it’s just used as marketing lingo to drive up sales. It’s hard to say the analogy to a cloud makes sense if you are still reliant on a single server’s hardware functioning correctly. Due to this, we wanted to offer a true cloud service with the reliability that you would expect from one. We also wanted it to be included in our standard server offering. And finally, we wanted to do all this without it being prohibitively expensive, which we think we’ve achieved!
Of course it depends on the configuration of the server, licencing, and any other factors, but choosing to host applications on our HCI VCS services can be as little as a 50% increase over a service without that HCI protection. Not only does this make meeting uptime requirements a walk in the park, but the peace of mind it offers is priceless. With similar offerings with other providers still costing more than double the original infrastructure costs, it’s a no-brainer from a business perspective.
The assumed reliability of "cloud" services
As mentioned, the term “cloud” computing gets used a lot. That’s not too surprising - as it inspires imagery of computing without being tied to hardware or the risks associated with that. And when people think about big-name cloud providers such as AWS (Amazon Web Services) or Microsoft’s Azure platform, it’s fairly common for them to assume that they will get a form of HA protection with their services as a standard. Unfortunately, this just isn’t the case.
With Virtual Machines (VMs) running on either AWS or Azure, you don’t get redundancy or automatic failover as part of their regular services. To get an equivalent service to our Hyper-Converged environment you would at least need to set up multiple servers, configure those to make sure they are kept in sync, as well as set up some form of load-balancing in front of them. That’s a lot of extra work and systems, with the costs bearing that out. Whilst you could look at setting up secondary servers in regions with slightly cheaper ongoing costs, it doesn’t add up to anything substantial.
There are alternate ways of getting that reliability too, such as running a similar VMware environment on AWS or Azure infrastructure. However, those are even more cost prohibitive, basically making them only a viable option for larger corporates with equally large budgets.
This is why, with many regular people already assuming that “cloud” basically means “High Availability” anyway, we figured why not make that true for all our VCS services. If people were assuming it, then our clients would be able to get the services they were expecting right from the get-go. However, we didn’t just stop there - we also saw an opportunity to provide that same protection for our entry-level hosting services as well.
A better kind of Shared Web Hosting service
Whilst it may not be surprising that we go to the utmost lengths to make sure our high-end services are protected, it might surprise you to know that we do the same for our cheapest services too.
In our industry, it’s fairly common to offer basic web hosting plans for businesses that need to host their websites or email accounts. Some clients can’t afford to pay for an entire server just for themselves or don’t need the horsepower that comes with it. What isn’t known though, is that many providers have their hosting plans make use of older and slower hardware to make them more economical. Not only that, but they almost never have much protection against the possibility of hardware issues causing downtime. In the end they usually aim to fill a cheaper server with as many accounts as possible, all to wring every last drop of profit they can get out of them.
Here at Micron21, we don’t play that game. Instead we offer a much more reliable service, with better support, and all at reasonable rates. To achieve this, we run our Shared Web Hosting off of our HCI VMware Cloud Server platform. Yes, you heard that right, customers on our entry level hosting plans benefit from the same protections against hardware-fault caused outages that our top-end clients do. This platform is also built using enterprise-grade Dell hardware. So for a few bucks extra a month, that’s a pretty good deal!
Going the extra mile time and time again
In everything we do, when presented with a choice between the best option and a cheaper work-around, we here at Micron21, always choose the former. In that spirit, we’ve created the most reliable virtualisation platform in the world by leveraging VMware’s Hyper-Converged Infrastructure technology. With our VMware-powered Public Cloud, we host mission-critical systems for our clients, giving them peace of mind that their systems are protected with guaranteed 100% uptime.
This protection not only covers the infrastructure that powers and cools systems with our Tier IV certification, but also the servers themselves via being hosted on a true cloud based platform. With those extra protections that are in place and our 100% uptime, we hope we’ve met the criteria required for being added into the nomenclature about other things that are “guaranteed in life”!
Interested to know more about our Hyper-Converged Infrastructure?
If you would like to know more, reach out to us on 1300 769 972 (Option #1) or email us at sales@micron21.com and we will be able to answer any questions that you have!