The hybrid cloud management (HCM) market has long been stagnant – but recently, that’s all changed. Companies are starting to solve cloud management challenges, and more than half of infrastructure decision makers at enterprises that have implemented the cloud are actively monitoring usage and creating formal cloud policies.
Too many enterprises conclude that they must purchase an HCM tool to tackle their cloud management and governance woes. For many, though, the multicloud management burden hasn’t yet become painful enough to justify the outlay. Before digging into the options available from HCM suppliers, first question whether HCM really is the solution to your problem.
In the early days of this market, the term “Generation 1” described products repurposed for HCM rather than built specifically to serve this purpose. Examples include BMC Cloud Lifecycle Management, IBM Cloud Orchestrator and Micro Focus Cloud System Automation (CSA).
Today, the lines are blurred, with these Generation 1 providers positioning other products, building radical updates and coming out with yet more new generations of products.
More recently, the conversation has centred around three core topics: breadth/depth of functionality for complex hybrid cloud use cases versus clean usability; full-feature depth versus composability to best-in-class alternatives; and integration to external tools via application programming interfaces (APIs) versus plugins.
When considering the generation of the tool, look at feature depth/breadth, composability and integration. From a depth/breadth perspective, bulky Generation 1 products aren’t simple – to both their detriment and their value.
These products have undergone significant modernisation and improvement while keeping the depth and breadth of their capabilities. They do this to serve complex use cases looking to manage a vast array of hypervisors and cloud platforms.
Highly complex enterprises and service providers often seek out Generation 1 tools for their granular governance capabilities and depth of support for on-premise platforms. Other solutions use a clean interface and speed-to-implement to differentiate.
If you have a simple use case using common technologies, a lighter tool will satisfy your needs while providing a better overall experience.
With regards to composability, all-inclusive tools deliver a ton of functionality, which simplifies integration and allows one tool to solve many problems. However, this also locks you into the technology at multiple levels.
Enterprises now seek composability for independence, standardisation across solutions and reliance on best-in-class for each segment of functionality.
Suppliers lacking capabilities heavily depend on the ability to supplement their lack of functionality by highlighting their composability stance and not reinventing the wheel.
More seasoned suppliers provide a “user’s choice” solution within the product, with the ability to swap out that functionality with alternatives.
Incorporating or integrating with popular tools, such as Ansible, Chef, Helm, Kubernetes, Puppet, Splunk or Terraform, is common among today’s HCM providers.
Building out integrations
Suppliers invest in building out integrations for their customers, either by demand or by necessity, and suppliers lacking core functionality may choose to supplement their services through integration.
Other integrations are due to requests, the need to stay competitive or through plugins developed by third parties or partners.
However, in addition to pre-built plugins, the majority of HCM tools publicly document most of their APIs so developers or systems administrators can build out their own integrations. Increased pressure from users has led many suppliers to fully publish all APIs in use — with no exceptions. Building out integrations can be time-intensive if you don’t have additional support, but it may deliver connections faster than your provider can.
However, this speed comes at an ongoing cost. It leaves the integration vulnerable to change by either party with little to no warning, so the question is whether waiting for an integration is worth the pain of not having to maintain these integrations over time.
Pick your management approach
Two differing mission statements exist for cloud management: make the underlying clouds invisible; or make the management of cloud resources invisible to cloud users.
Your HCM supplier may limit this choice, as the “invisible cloud” concept gives its platforms more power as the single control point for provisioning new resources for both managers and developers.
The tool gets out in front of cloud usage to create one experience and set of rules across cloud platforms – but this lacks a developer-centric mindset. It dictates that developers use this experience to launch new resources, but developers want to pick their own experiences and tools.
Many specialists have found “invisible management” the more realistic mission statement, especially in developer-centric organisations.
Most HCM suppliers make “invisible clouds” the default option, with the ability to bypass this option if “invisible management” is the preferred approach.
In practice, providing invisible clouds translates to abstraction from the native APIs. The latest approach? Create an API layer that can intake data sources from different sets of APIs and formats.
Some HCM solutions use this same API layer to translate their instructions into the native API requests of that provider, often providing a “shim” layer that includes any specific compliance rules or directions that the provider dictates.
Despite the shim layer, the resources provisioned are standardised on the native platform APIs of the cloud platform in use. If you discontinue your HCM contract, it’s just the shim of policies that goes away, whereas onboarding existing resources requires a template conversion to the HCM format.
Looking at invisible cloud management, the seemingly simple task of allowing provisioning through a different platform Is tricky.
Most HCM tools want to be both management and developer dashboards, and using a different experience requires either exporting the templates to the other platforms; immediate conversion of a native template to their own postmortem; or maintaining the native template but providing after-the-fact remediation if policies are violated.
The less intrusive method, after-the-fact remediation, creates a delay to compliance and can involve a long list of manual tasks – defeating the initial productivity gains the alternative access sought. More intrusive methods, such as workload conversion, are often done to ensure the same actions can be taken on the workload as those provisioned directly through the HCM portal.
The era of pragmatic cloud is upon us. It doesn’t matter which tool looks nice or which supplier has momentum – ultimately, you need a solution that will work for your organisation.
Indeed, you’re not tasked with picking the best hybrid cloud management product in the world. Certain features will likely make a significant difference to your organisation, whereas others fall into the nice-to-have category.
Some options simply cost too much. Every cloud platform, tool and service is in competition for budget, so when it comes to cloud management, you may find a lightweight solution with just enough capabilities – one which better serves the budget for your company’s other needs.
Lauren Nelson is a principal analyst at Forrester. This article is based on an excerpt of Forrester’s “I&O pro’s buying guide for hybrid cloud” report.