DevOps & SDDC: Perfect Marriage for a successful Cloud era?

DevOps is a system deployment methodology that organizations can use to improve the speed and quality of their project deployments. DevOps has started a revolution for traditional IT shops where practices and tools have been somehow affected by previous attempts to do things right, finally under DevOps it might be possible to achieve them since day 1.

On the other hand, Software-Defined Data Center (SDDC) brings alternatives to the regular Server, Desktop and Application Virtualization paradigm, by putting together not only Storage (SDS) and Networking (SDN) Virtualization but also pushing to automate Data Center operations inside a Cloud implementation, either Private, Hybrid or Public Cloud.

DevOps allows SDDC to become mainstream, programmable and scalable, by putting together the proper DevOps methodology applied to Software-Defined technologies (and APIs) that have been already successfully implemented and tested in Labs, Proof-of-Value (PoV) and small/medium implementations. At this point, I do not personally foresee any major Cloud implementation without deploying both techniques first: DevOps and SDDC, all together. There is no other way around to automate regular Cloud operations and achieve the long-term business goals set by Cloud Service Providers (CSP) and Managed Services Providers (MSP) that will, by the way, finally coexist in the same market ruled by collaboration and messaging between multiple APIs. This is the only way to grow the business and scale out/up according to realistic Customer expectations.

Going back to DevOps principles, this term was originally being used to describe a moderator between the two groups who functions as a type of scrum master to help developers and operations teams keep application lifecycle management (ALM) top-of-mind. the ALM will be directly impacted by SDDC in terms of all infrastructure requirements will be satisfied after running automation scripts inside the DevOps cycle. As result, the new “Cloud DevOps” will be useful to describe the upcoming relationships (and frictions) between a) Cloud Architects Team, b) the Infrastructure Team composed by Virtualization, OS Sysadmin, Storage and Networking experts who will also act as quality assurance for infrastructure purposes, plus c) the Platform/Application Team in charge of core PaaS/SaaS/DBaaS related activities inside the Organization. DevOps is the only way to succeed in the SDDC-powered Cloud 3.0 era, by tearing down silos between all these traditionally divorced and isolated functional groups: Architecture, Infrastructure & Developers.

Finally, DevOps & SDDC marriage will push to reinvent ITIL and COBIT practice and requirements, from the business prospective DevOps & SDDC will reevaluate the traditional way to meet SLAs into a more streamlined exercise. It will bring such dynamics to the table that finally Upper Management will dramatically enforce similar integrations to different business needs, with or without having ITIL/COBIT in place. Flexibility and Business needs will rule instead.

As a reference, Techtarget states “The DevOps toolkit includes configuration management tools like Puppet and Chef, a repository like GitHub for storing versions of code, indexing tools like Splunk, tools for monitoring how changes to code affect the environment like Nagios and scripting languages like Perl, Python and JavaScript.” This represents a golden opportunity for all of us, Cloud architects, to revisit automation tools available for each Cloud Platform and advice best option in each case. As part of the Data Center Diversity, we might be open and willing to support VMware & Hyper-V for Mode 1 Applications (i.e. ERP, Business Critical Applications) and OpenStack & KVM for Mode 2 Applications (a.k.a. Cloud-native Applications).

Let’s think DevOps as a disruptive movement and approach, as long as it implies the real integration with SDDC to finally impact positively the Cloud adoption for years to come.