Websites are a vital arm of your business and it needs to run smoothly and swiftly. Other developers do not provide any extras. In most cases, they simply provide the website and expect you to purchase hosting separately and then manage it yourself. This causes many issues as hosting is not something that you can easily learn. It takes years of experience to be able to know where issues could arise and how to fix them correctly. Hosts have a very bad habit of blaming issues on everything except their own system which causes a back and forth between the developer and the host. We host everything on our server so that should any issue arise, we can address it and fix it as we have full root access to every single area of the server.
In the IT industry, there is a term which is “SPOF” which stands for Single Point of Failure. A single point of failure is an item that will cause everything else to fail. For example, with a vehicle, a single point of failure is one wheel. If that wheel fails, then the vehicle can not operate properly. Our systems are setup to ensure that there are as few single points of failure as is possible. In many ways our systems are like aeroplanes, which use a double redundancy configuration. Should one item fail such as lowering the wheels when coming in to land, there is a backup system that will lower the wheels, thus preventing a single point of failure.
When it comes to your website, we do everything we can to avoid issues or crashes. Much like an aeroplane, we deploy a minimum of double redundancy everywhere that we can. This ensures that if one item fails, there will be no disruption while the item is addressed and corrected.
Below are the key areas which have a minimum of double redundancy to ensure a smooth operation.
- Load balanced connection
- 4 backbone connections including Level 3, Internap, AboveNet, and TW Telecom. A backbone is the spine of the internet with connections to internet service providers.
- Each connection enters the datacentre in a completely different location so that even if an issue effected part of the building, it could only effect one connection and your website would remain fully operational.
- Two completely separate internal networks, both of which connect to all websites. This ensures that any network issues are avoided as a secondary network is always standing by.
- The network is a Service-Oriented Network Architecture by the expects at Cisco.
- We only use the best equipment possible from companies with sterling reputations. This ensures the best possible quality and stability from our internal network as well as providing two access methods to our load balanced internet connection.
- All servers, networks, internet connections and hardware is connected to an Uninterruptible power supply. If there is a power failure, the Uninterruptible power supply immediately provides power, ensuring that there is no interruption.
- Four (4) on site autonomous DC plants, providing power should there be a need for extended backup power. Each DC plant is in a diversely different location so that an issue effecting one DC plant could not effect the other DC plants.
- Six (6) Diesel generators with auto start and switch, ensuring that when power is needed, it is provided within seconds. In the meantime, the Uninterruptible power supplies and DC plant will provide emergency power, ensuring that there is no interruption. The generator provides enough power for the entire building as well as having an extra redundancy to ensure there is power when needed.
- Fuel storage for at least 24 hours with a contracted guarantee for extra fuel delivery within 2 hours.
Servers create a lot of heat and need to be cooled to ensure smooth operation.
- Air Conditioning with humidity control.
- Under floor cooling to ensure the server units are also cooled from the bottom.
- Pre-action sprinkler rated for telecommunication equipment/computer room
- Integrated smoke/heat detector system
- Our datacentre is one of the most secure available and includes biometric scanners, 24×7 security cameras, card readers and key pads.
- Server rooms are secured with card readers and key pads.
- Server cages are locked.
- Only certain staff have access to servers, server rooms and server cages. Each member of staff has a unique code which is logged when they access so that we can easily see who had access and when.
- Our servers use the latest software which ensures that any known security issues are patched as fast as possible.
- Automated anti virus scans are completed monthly to ensure a clean system. Viruses on servers are extremely rare and very very few servers have any anti virus systems.
- Full firewall and monitoring system prevents users from port scanning, brute force hacking and other similar attacks.
- Passwords are randomly generated and have a total of 328 decillion combinations. A decillon is 10 to the power of 33, so that is 10 with 33 zeros after. 328 decillion would be written like this: 3,280,000,000,000,000,000,000,000,000,000,000,000. That is the total possible password combinations required to guess just one of our passwords! Using a very very high powered computer, you could make 17 billion password guesses per hour. This would take at least 398,144,091,233,324,000,000,000 days to guess all password combinations. That equates to almost 10 Sextillion years (10 trillion billion years)! If someone were to attempt to guess one of our passwords, our server security would block them after 3 attempts. Their 4th attempt onward would be ignored by our server and they would not be able to connect to our server as the connection would be blocked by our security system.
- Daily backups, stored on a hard drive dedicated for backups. Most companies store backups on the same hard drive that they use for websites. The hard drive used for websites is usually very busy and should it fail, you lose all backups. We make sure that the backups are on a separate hard drive so that they are secure. Should one hard drive fail, we will have a copy of your website on the other hard drive. In the unlikely event that both hard drives fail together, we use a secondary backup which is detailed in the next bullet point.
- Daily database backups, stored locally and also stored on a Google Cloud Server to provide a minimum of double redundancy.
We use two systems to ensure maximum stability.
- Internal system monitors the server and has the ability to correct errors or issues. If needed, it can restart the server to resolve the issue.
- External system connects to the servers every 5 minutes, 24 hours a day, totalling 288 checks every day. The checks are completed from a huge bank of global servers so that any issues connecting from a certain part of the world can be identified and addressed. When any issue is found, we are emailed and sent an sms so that we can immediately diagnose the issue and correct it.
Hardware and Software
- We use only the latest and most stable hardware and software. All software is updated on a monthly basis to ensure that the server is fast, stable. Hardware is regularly checked to ensure it is error free and is operating as required.
- Speed is absolutely vital and we make sure all websites load at optimal speed.
Content Delivery Network
We deploy a content delivery network for premium websites. The network provides images and files from a server that is closest to the end user. This reduces the latency and keeps the loading times at a minimum.
As you can see, we are utterly serious about what we do. We are the best at what we do because we have the best tools, hardware and software as well as working with the best companies in the business to provide the best possible solution for your business. Our mission is to be the best at everything that we do, while providing the best possible solutions for your business.
Please contact us to talk further about what we can do for your business.