Is the cloud resilient enough to keep your services up and running in the event of an outage? That’s a good question for companies to ask themselves as they move more and more of their infrastructure and services to the cloud.
There’s no doubt that the cloud is one of the most prevalent trends in computing today. Microsoft stats show that 30% of SMBs currently use some cloud services, with another 48% expecting to add cloud services in the next 2 to 3 years. With cloud computing becoming much more mainstream among the SMB crowd, it’s now time to make sure that the cloud can support an organization’s computing needs and service requirements in the event that something goes wrong. And unfortunately, as it stands right now, the cloud might not be up to the task of hosting a company’s critical services.
One problem is that SMBs figure there is safety in numbers. If they throw more and more services into the cloud, and if more and more companies are also doing so, there must be a measure of protection against service outages, right? Unfortunately, these companies are not taking into account that there are weak links that still exist.
One major weak link today is how users configure virtual machines. If these machines are not configured correctly, it will result in errors that have a ripple effect on services that run on the machine. The growing number of machines – both physical and virtual – within organizations is contributing to a greater number of configuration errors. Quite simply, with more machines to configure, there is more that can go wrong.
To combat this issue, companies should take a few steps:
- Automate more and more tasks in order to keep up with the growth in machines;
- Employ application-level monitoring, which will help alert organizations to application availability degradation; and
- Not depend completely on just one provider of cloud services. In the foreseeable future, it would be wise for firms to run their high availability services locally, while using the cloud for disaster recovery.
Even if companies tackle these issues, a self-healing cloud is still five years away. It’s not enough for organizations to just throw their applications into the cloud and expect everything will be safe. Just as with a physical infrastructure, it requires planning and assessment to make sure any cloud deployment fits the company’s needs and service requirements.
Cloud image by basketman