We've all know how security is among the top - if not actually the top - concern for CIOs when making decisions about moving workloads into the cloud. Less often spoken of is visibility: the ability to view in detail and to manage those assets when they're parked on top of someone else's infrastructure - whether that infrastructure is simply a datacentre facility or a hosted server.
Criticality of visibility
Without visibility into your data, it's hard if not impossible to troubleshoot, to resolve governance issues, and to ensure that security controls are actually working.
For example, enterprise cloud users can find it difficult to monitor data in motion, such as when it's moving through a virtual switch. When data is at rest, it can be just as difficult to locate, as data will be stored in different pools across multiple servers and locations.
So visibility has to be one of the key criteria for assessing the compatibility of a cloud provider for your applications, policies and procedures.
Ideally, you want a cloud monitoring solution to monitor, diagnose, alert, and notify you of problems in your cloud infrastructure. It should provide platform metrics with high levels of granularity, using your choice of messaging protocol, whether SMS, email, or webhook.
It should help you diagnose and get support when an issue adversely affects your resources, while informing you about their current and past health statuses and helping you mitigate issues.
Data and resource management should not be an add-on but a given: this is nothing less then you would expect from your own infrastructure. Monitoring capabilities should be available without any need for manual configuration or purchase of additional tools.
This means providing you with monitoring telemetry including platform- and service-level metrics and logs, enabling you to configure alerts to take intelligent actions on that data, and so empowering you to unlock deeper insights and analytics on top of the telemetry.
Those configurable alerts are key. Alerts based on specific metrics flag up issues and - if you've picked your cloud provider carefully - allow you to automate the action to be taken based on logged events such as VM reboot, deployment failure, or user permission change.
A monitoring solution should integrate with your existing monitoring solutions, using APIs to enable the management of the cloud- and on-premises-based infrastructure from end to end, al from a single pane of glass.
The visibility business advantage
From a business perspective, this all makes perfect sense in terms of reducing learning curves and nipping problems in the bud before they become major, revenue-affecting calamities.
But monitoring can also shade into analytics, which can add business value. This means allowing you to detect trends in application performance and behaviour, identify usage patterns, and answers to questions about website performance.
So cloud systems management is not just about keeping en eye on systems and responding to alerts. Like any form of predictive maintenance, as used increasingly by companies in all segments, it's about enabling the ability to become more customer-focused, more agile and flexible, and cutting the costs of managing infrastructure.
If a cloud provider is not delivering on this front, they're not adding enough value.