Earlier this year cloud computing provided by Amazon Web Services experienced a bit of downtime. When looked at overall, the amount of time their systems were down actually are rather minuscule, but that’s not the point. What is salient is that many of the Internet’s leading websites, social communities and other online businesses rely wholly or in part on the computing power, storage and other processing capabilities provided by Amazon’s cloud services.
When active users of sites like Reddit, HootSuite, Foursquare, and many others including CRM blogs are suddenly, with no warning, unable to access their accounts due to circumstances beyond the site’s control, things can get a bit hairy. For a small percentage of AWS clients, real data was lost in the outage. Now there is permanent damage that has to be reconciled for something the service providing company had no part in causing.
To be clear, the advantages of using cloud services far outweigh the potential for challenges of hosting ones systems in the cloud. Most cloud providers have in place a number of automated back-up features which theoretically maintain data integrity with real time mirroring of client information. Transactions which rely on access to historical data can, again, in theory, be back up and running in moments when things are done correct in the cloud.
So what are those things a company needs to do right when employing the cost effective computer and storage services in the cloud? How much is enough to protect company and customer records from being lost permanently? Should all hope be lost and the business simply invest in hardware and applications completely within their own control?
From the perspective of re-budgeting ones online operations to allow for investment in company owned equipment, only the business can answer that question. Most companies decided to use cloud computing services precisely because the entry costs are extremely reasonable. To shift, mid-stream, and start hosting and serving applications and services from ones own data centers simply may not be feasible.
Perhaps the better solution is to spend more time looking into backup options that spread the risk across more than one company or network. Some of the victims of Amazon’s recent cloud crash complained they understood their data was backed up, automatically and regularly. If this was the case, as the complaints go, why then did some of AWS clients suffer permanent data loss?
The smarter way to go is to understand that these things happen. As such, finding a service provider not affiliated with Amazon to perform the backup tasks required by the business is important. Better backup solutions will have redundant systems to mirror client data across a number of physical locations.
But this may not be enough peace of mind considering recent events. In addition to securing services to mirror information, the business might want to invest in a local repository as well. Depending upon the volume of data under consideration, it might be reasonable to as a tertiary position, to have data streamed into onsite storage in addition to the other methods currently being deployed.
Computers are no more infallible than any other machinery. Taking the time to secure ones data, as well as its back up, in near real time can be one of the most important expenditures a company makes. Understanding that certain scenarios are going to happen, sometime, to some businesses can help smooth over the wrinkles a massive cloud computing crash causes. If the business has adequate protection against valuable data loss, the ability to get back up and running as soon as the cloud is flying high again proves to its customers it knows what is happening, how to prepare, and bounce back.
In the end, these are the types of companies customers want to deal with. If the business provides logical, heartfelt explanations about the situation as well as getting over the hump caused externally, then even more customer respect tends to flow towards the business as well.
