Part 4 of the Rise of Platform Operations Series, for Morpheus Data by Jason English
How hybrid cloud platform operations can provide an evolutionary jump in how the enterprise thinks about IT service delivery
“To dream the impossible dream
To fight the unbeatable foe
To bear with unbearable sorrow
To run where the brave dare not go…”
– “The Impossible Dream (The Quest)” from Man of La Mancha
Enterprise IT leaders have long dreamt of delivering self-service application environments that are easy to maintain and monitor within budget constraints, yet are as agile as the DevOps teams they support.
Perhaps this quest is no longer as quixotic as it used to be, thanks to the advance of elastic computing resources atop hybrid cloud platforms and hyperscalers, driven by cloud-native technology, infrastructure-as-code, and delivery automation.
Even if dev and ops teams sally forth with the bravest of automated cloud provisioning efforts, without a good connection to the business and organizational realities represented within service management, observability and FinOps platforms, they might as well be tilting at windmills.
There’s a lot of excitement today around platform engineering, and moving away from classic, sequential delivery processes by setting up self-service provisioning atop hybrid cloud infrastructure.
By comparison, hybrid cloud platform operations sounds less sexy, though it is equally as important as its counterpart. We want to offer self-service, yes, but we also need to monitor and unify the ongoing operations of the cloud resources we spin up today into the future, in a manner that is embedded into the everyday process engineers participate in.
Embedding service management
in modern practices
In the classic model of provisioning, a centralized IT Ops team took in dev team requests from a system like ServiceNow or Jira as a front door.
Each request ticket—hopefully, accompanied by thorough documentation—would then go through a number of resource, budgeting, quality control, and security approval processes and system checks from there, until the Ops team finally built and deployed the requested system a few weeks later.
Modern development teams can ill afford to wait for such cycles within tight timelines. Self-service automation is a core tenet of the DevOps movement, and without it, provisioning and maintaining environments will become the primary bottleneck to productivity.
Unless a course correction is made quickly, teams will ‘go rogue’ and do their own thing without the blessing of IT. In fact, many frustrated teams did just that, which resulted in a rash outbreak of ‘shadow IT’ over the last several years. Ungoverned cloud usage can cause even worse problems when cloud costs spin out of control and deployed applications are breached or fall down in front of customers.
Internal developer platforms are a more modern way to address the need for speed with safety, though we must deal with the underlying reality of populating and updating configuration management databases and tools, and interfacing with ITSM systems that have become critical for servicing customer needs.
ITSM tools and platforms still have an important role to play in a hybrid cloud platform operations framework. Now they are simply a layer of the cake, baked into the system rather than bogging it down…
Read the entire BrainBlog here.
