First steps here should be the same as if a company were using a traditional “client server” application environment, where a physical server is in place to literally ‘serve’ the requirements of ‘clients’ in the shape of physical desktop and mobile machines.
So this means that initial User Acceptance Testing, or UAT as it is known, still needs to be carried out to ensure that your stakeholders (or “staff” if you prefer the more common term!) can tell the IT department whether the cloud applications in place meet their needs, or not.
Cloud watchdogs on patrol
From this point, a company that has chosen to move to hosted cloud-based applications and storage will also want to at least consider the option to use a third party testing service. These ‘watchdog’ type systems are designed to report on metrics such as input/output times (commonly written as I/O) so that IT can monitor the cloud to see if there are any “issues” to be aware of
A cloud watchdog service such as Gomez or monitis will also provide stats on general response times and application performance. This information is crucial for the customer’s IT department (or IT service consultancy), as it will enable them to fine tune the cloud infrastructure that the service currently rests upon. Monitoring provides this visibility; and from visibility comes control and manageability; and from control comes profitability.
Raising the alarm
In practical terms, that customer who wants to impose higher levels of diagnostics and analysis upon their cloud computing service will want to set alarms. These alarms will warn the firm’s IT function if cloud application performance or wider service levels drop to a point where they become unacceptable.
Cloud application monitoring can be carried out at multiple levels. While one customer might be happy with summary performance reports once a week; another customer might demand monitoring data on the performance of multi-step transactions across several co-connected cloud-based resources every 60 minutes (or less!) on a 24-hour basis.
You don’t need to understand multi-step transactions and co-connected clouds to grasp the fact that monitoring can be high-level or focused on deep root cause analysis.
Whichever level of cloud monitoring is put in place, the customer will ultimately “view” the monitoring report in what is likely to be a dashboard-style presentation. This set of gauges and digital instruments is an extremely user friendly way of getting a quick visual impression of what is happening in a cloud-based application and storage system.
The same approach is used by Business Intelligence (BI) system vendors, who may also include cloud monitoring options in their BI controls. As more companies start to adopt the cloud, monitoring tools will no doubt become more sophisticated -- now is the time to at least research this option as part of your cloud adoption policy if nothing else.