For any business based on a cloud platform, being prepared for an outage is absolutely crucial. At ClicksMob we are fully focused on delivering the best service to our clients and partners, and this means being ready for unforeseen technical issues. This necessity to be prepared was demonstrated late last month when Amazon Web Services (AWS) suffered a major disruption.
Across the world, some of the internet’s biggest websites and apps, including Tinder, Airbnb and IMDb suffered serious disruption when more than 20 services on the AWS platform began failing. Over the course of eight hours, users reported problems accessing content on a huge number of websites. This is likely to be a very costly mistake for Amazon. Back in 2013, a similar outage cost Amazon a reported $1,100 per second.
Due to forward planning and the strongest commitment to customer service, ClicksMob was able to have our platform up and running and fully operational just 20 minutes after initial problems were reported.
So that we are prepared for system failures such as the recent AWS outage, our system is composed of several sub-systems that are able to run at any location within the Amazon domain. This means that we can mobilize our sub-systems across regions in order to provide a system recovery within a matter of minutes.
Here’s the technical bit: Because these sub-systems are regional and self-contained, we can very quickly identify a problem such as Amazon’s outage that specifically effected the N. Virginia region, and quickly start an alternative EC2 instance at another region in order to immediately return to service. During the startup process, each sub-system performs a validity check on the data and automatically rebuilds any missing elements so that the experience of customers and partners is not affected.
We ensure that our data-center is constantly backed-up so that no information is ever lost, and the platform can be quickly restored without the loss of any work. In fact, in order to provide an even better experience for our customers, we are currently working on a solution that will enable us to operate multiple data-centers that are constantly synced and backed-up with the master data-center.
The ClicksMob development and delivery environment is based on an Application Lifecycle Manager (ALM) approach, with Continuous Integration (CI) and Continuous Delivery (CD) solutions. This enables us to very quickly deploy the latest code to all of the ClicksMob sub-systems with the push of a button.
We place customer support and service as the highest priority at ClicksMob so we were delighted that all our planning for situations such as this came good and we were able to get our services up and running in such quick time.