Cloud service outages at companies such as Amazon and Microsoft receive vast media attention considering the impact on their vast customer base and reputation as tech giants. The problem of cloud IT service outages has expanded to such an extent that customers hate raising this issue to their vendors who offer inadequate cloud monitoring tools and little support to offer.
These cloud outages have revealed several prominent problems with the cloud IT service structure and one of them is associated with the concept of cloud “hype bubble”. For years, the idea of a cloud as a solution to the majority of IT challenges has been propagated. “The cloud is easier to use, faster, cheaper, more adaptable, more versatile…”
Well it is, but it has to be made as such, and it has to be maintained as such.
The reality is that certain companies, as well as individuals, are using clouds for the wrong reasons – to bypass standard problems of traditional IT. Needless to say, the cloud systems use the same hardware as the traditional IT, and just like traditional IT, it relies on software, it needs management, it utilizes processes, etc. Although it is impossible to prevent outages, the consequences of even the largest outages can be significantly reduced with the appropriate cloud monitoring tool.
The Costs of Outages
The cost of even a short outage can be extreme. Google’s 5-minute outage from 2012 cost the company over $500k. Below are bullet points which help illustrate how much data center outages cost.
- Companies lose over $160,000 per hour of downtime and the cost rises constantly.
- 59% of Fortune 500 companies experience a minimum of 7 hours per month of downtime, which results in a loss of $46 million per year.
- Average recovery time after an outage is 7.5 hours.
- 1% of outages last more than 12 hours and additional 22.7% of outages last between 8 and 12 hours.
- The average company experiences at least 1 large scale cloud outage and 5 partial outages per year.
Flaws of The Native Cloud Monitoring Tools
The monitoring tools provided by the IaaS providers are painfully inadequate. Users are limited to minimal metrics, and closer analysis of the problems that occur during the use of applications is not available. Since business customers typically lack expertise for 24/7 monitoring with relatively complicated native tools on the platforms, resolving the problems takes extra time and costs more money.
A great example of inefficiency is AWS CloudWatch, the native tool for the largest cloud provider. AWS CloudWatch has limited functionality. It only holds on to data for 2 weeks, which prevents resource planning, there is no 24/7 alarm watch, it cannot inspect metrics specific to your application, and the list goes on.
Azure’s cloud monitoring is not great either. There is nobody to wake up your team if there is a problem in the middle of the night. No one is keeping an eye on your cloud environment. Also, glitches can be a problem, as well as the slow data flow.
HOSTING Cloud Monitoring Solutions
All HOSTING-managed solutions come with our Unified Monitoring solution which includes 24/7 monitoring and incident response from your support team. Our platform is built on top of ScienceLogic – the world’s most comprehensive IT asset monitoring software. HOSTING can monitor every component of your environment, from dedicated servers to PaaS services and everything in between, including practically every asset you have at AWS, Azure, and other hyper-scale cloud platforms. Customers can also leverage our monitoring platform to track, alarm, and review the performance, capacity, and availability of their custom application components – all through one pane of glass managed by HOSTING.
With a high-end cloud monitoring tool, your company can save thousands of dollars and avoid the wrath of unsatisfied customers. Want to learn more about cutting your losses? Download our white paper now.