Getting Started with Hybrid Cloud Management

For many enterprises, hybrid cloud hosting can create the right mix between the savings and efficiencies of cloud and the compliance, performance and other requirements that are either best-served or which need to be served by on-premises infrastructure. The question today is how to manage the assets you decide upon in the most effective manner?

Any quality data center – whether local or remote – will have operations and tools in place to look after the basic care and service of the operating system: patching, anti-virus protection, as well as monitoring for storage and processor capacity. There’ll be a responsive 24/7 NOC (network operations center) that can troubleshoot and react if things go wrong. And finally, there’ll be auditing – a critical component of IT infrastructure, especially in an age where compliance requirements are pervasive across most industries.

The management tools to accomplish these functions – and the considerations for selecting them – have changed with the advent of cloud, however, and our choice of tools is a key part of developing a hybrid management strategy.

The Traditional Tools

Traditional, on-premises, data center management tools are often agent-based. This means you’re installing an agent into the operating system, whether it’s on a physical server or a virtual machine, to provide a particular type of functionality. However, what these agents also do is consume resources on that machine, and they may introduce incompatibilities with certain applications.

Traditional tools are also often designed for a LAN environment, assuming ubiquitous connectivity, low latency, and effectively unlimited bandwidth. In turn, the protocols that run between the agent and the management platform that often gets installed alongside can be quite chatty. Although internet connectivity has dramatically improved over the years, extending these tools into a hosted environment that operates across geographical distance and higher network latency may not yield the desired result, and will consume limited WAN bandwidth.

Second, these tools often require administrative level access to the instance being managed. When working with a service provider, the clearer you can delineate who ultimately has administrative rights over machines, the easier that relationship will be. If both the enterprise staff and the service provider have administrative access to systems, it can be difficult in the SLA to define the boundaries and accountability if there is an outage or an incident.

The Cloud Tools

In contrast to the traditional ones, cloud-enabled tools for data center services have changed how we’re able to provide some of these basic management services. In particular, if utilizing tools that are able to work through a hypervisor owned by the service provider, then that delineation can be made much clearer and often any need for administrative access is eliminated.

For the management of hybrid environments, working through the hypervisor allows service providers and in-house teams to work more efficiently by, for example:

• Reducing the amount of data sent between networks when creating backups
• Easily monitoring, managing and collecting metrics at the operating system level
• Auditing operating systems with cloud tools without compromising control

Mix & Match: Hybrid Tools

For any given enterprise, at the beginning of their journey towards a hybrid environment, the tools in use by the in-house teams might be quite different from those being provided by a cloud services provider. Often the tools used in the enterprise were selected because they provide a particular value or a particular customization that is important for the existing systems.

On the other hand, the tools chosen by service providers are often selected because they operate efficiently at scale and provide the ability to deliver a standard offering. In these instances, teams must work together to evaluate the business need for customization versus the ability to work with, or within, the tools provided by the service provider, and ensure that they can work towards an efficient delivery model at a compelling price point.

That said, in some instances, where, for example, in-house teams have spent years fine-tuning their monitoring environment to provide the right information in the right way, it may not be the most effective move to change to the provider’s solution.

Similarly, an enterprise might also have the credentials processing management in place for their own two-factor authentication. Here, it may make sense to actually extend the people, tools and processes in place into the hybrid cloud environment as well. That said, in a hybrid cloud environment there is also the need for management of the instances themselves, i.e. the provisioning and de-provisioning of virtual machines. As such, service providers would still need to provide strong authentication into the provisioning portal, which needs to be just as, if not more, secure than the underlying operating systems.

A skilled hosting service provider can accommodate almost any combination of the management tools that an enterprise needs to operate in ways that are both customized and cost-efficient. They will be able first, to provide counsel and the benefit of their experience as to which tools might work best. And second, will provide the expertise to assist with or manage the integration of whatever hybrid combination of tools provides the optimum solution for each particular customer.

This post originally appeared on the NaviSite blog.