The cloud represents a mass of moving parts. Multiple vendors, resources and assets may be intertwined to deliver a crucial service to users or customers. As cloud deployments expand, managing these multiple parts is beyond the capabilities of mere mortals and requires orchestration and automation to keep things working efficiently and cost-effectively.
“As organizations move more workloads to public, private and, increasingly, hybrid clouds, orchestration is a requirement,” writes cloud visionary David Linthicum, senior vice president with Cloud Technology Partners.
“The good news is that most enterprises are new to cloud orchestration and are still determining the tool sets that they need,” Linthicum continues. “That’s good news because trying to use a legacy IT orchestration tool for hybrid cloud — where well-defined APIs are a must — typically won’t work.”
Limitations of CMDBs
Many enterprises have spent years wrestling with data center configuration management databases (CMDBs) in an effort to catalog IT assets needed to support self-provisioning. That’s tough enough to do within the confines of the data center; but extending these tools to encompass dynamic cloud resources is another matter altogether.
“CMDBs are the systems of record for the enterprise data center, but they can also prove a major stumbling block to managing a successful cloud implementation,” asserts TechRepublic contributor Keith Townsend. He notes that each data center silo may have one or more CMDB, which must update a central CMDB. “Anyone who has performed any real-time extract, transform, and load (ETL) processes between two systems understands the pain of maintaining state between CMDBs with completely different database schemas and architectures.”
With more than 85% of enterprises employing multiple clouds, according to IDC, and relying on different cloud providers, automating and orchestrating IT assets becomes even more complicated.
“From an operational perspective, the main stumbling block appears to be the lack of effective tools to manage the data environment once it leaves the confines of the data center,” writes Arthur Cole for IT Business Edge. “And this becomes particularly worrisome when, as is often the case, data is not limited to a single provider but is divided among many. This is why cloud orchestration continues to be a primary, albeit elusive, goal for the enterprise.”
Meeting the Challenge
For enterprises already struggling to hire or retain IT skills, this can seem like yet another hurdle.
“It’s true that learning the ins and outs of the infrastructure and lingo of more than one cloud can be challenging, especially for a smaller company,” writes Rackspace CTO John Engates. “Meanwhile, bigger companies are faced with tremendous competition to retain the specialized engineers and architects who are versed in multi-cloud, meaning that even they often struggle to keep the required skills.”
But Engates advises that “this isn’t a burden that IT departments need to shoulder alone and it doesn’t need to get in the way of benefiting from multiple clouds.” Managed cloud providers can help enterprises fill those gaps with highly skilled teams proficient in using tools such as OpenStack Heat to help automate and standardize your cloud infrastructure. Click here to read more about OpenStack cloud orchestration.