You are currently on IBM Systems Media’s archival website. Click here to view our new website.

Bookmark and Share

Recent Posts

More on Ops: CloudOps and DevOps

October 29, 2018

With this post, I continue my focus on computer operations (Ops). Last week, I discussed SysOps, NetOps and WebOps. These are three important systems or specializations within computer operations. In this post, I explore CloudOps and DevOps, which are two different ways that Ops is engaged. Although they’re different, CloudOps and DevOps teams are doing what’s necessary to help create and support an innovative vision of IT and how it should operate in support of their organizations. Let’s explore these topics.

CloudOps Specialization  
CloudOps is about cloud operations. Why do we need CloudOps? Isn’t it just like commercial web hosting or in-house IT? It’s true that cloud computing grew out of legacy system and service approaches but it’s different, and it benefits from a specialized operations approach. I know this from experience as I was a developer commercial cloud offerings.
Cloud computing, or “the cloud” is focused on the delivery of on-demand computing resources like applications, often delivered as software as a service (SaaS), over the Internet. It’s more than just applications delivered on demand but SaaS is very popular with a 2018 revenue estimate of $71 billion.

Consider these three cloud differentiators:
  1. Scale up or down quickly and easily to meet demand. This is called elastic resources.
  2. Pay for what you use. This is called metered services.
  3. Get what you need. Most or all the IT resources you need have do-it-yourself access. This is called self service.
Supporting elastic resources is a significant challenge for CloudOps as there always needs to be compute, storage and network resources “in the bank” for use by customers. This must be a significant capacity planning headache.

Metered services are also mission critical for CloudOps. If the service isn’t managed in an excellent manner, cloud service provider revenue could be lost. Confidence in the cloud provider could also suffer as well as its reputation.

Self service has its own set of challenges. In addition to service availability, CloudOps is involved in user support. Self service, doesn’t always mean you don’t need help and support from time to time.
These three areas are just the beginning of the CloudOps challenges. There are also the regular IT management practices—security, change management, issue and problem handling, and more.  

Continuous Operations Links CloudOps to DevOps
CloudOps is a service that that lives in a world that seeks to provide continuous operations. Continuous operations targets the ability to run cloud-based systems in such a way that there's never the need to take part or all of an application out of service. This is something that we’ve seen fostered in SysOps on mainframes for many years, and due to the importance of IT in so many aspects of business, it has become an aspect of may other computer services. The continuous operations goal supported by CloudOps is related to continuous integration, delivery and deployment that are known collectively as CI/CD.
DevOps Is Growing Strong
DevOps is the joining of development and operations. It’s a software development approach or method that combines the creation of software with IT operations. So why do we care about this joining of forces?

We care because DevOps is a tool to shorten the systems development life cycle while also delivering function, features and fixes that closely align with business wants and needs. The DevOps approach also puts a focus on using automation for testing, building on the work that has been done in the past on maintaining test cases that are run automatically to check out changes. It’s interesting to note that regular test cases, as well as those used for stress testing, can also be the basis to ongoing monitoring of the system. Single assets used for multiple uses can be a powerful idea for saving an organization time and money.

The promise of DevOps is that it brings together developers and system administrators to work toward a common goal. The degree of success is largely influenced by leadership and organization commitment not technical matters. IBM has been using DevOps internally and that has enabled it to make good progress around acceleration of delivery cycles. Highlights from the most recent 2 to 3 years include:
  • Project initiation: 10 days to 2 days (8 days faster)
  • Overall time to development: 55 days to 3 days (52 days faster)
  • Build verification test availability: 18 hours to < 1 hour (17 hours faster)
  • Overall time to production: 3 days to 2 day (1 day faster)
  • Time between releases: 12 months to 3 months (9 months faster)

What’s Next?
In the next post, I’ll discuss DevOps and ITIL. Does one eliminate the need for the other? How do they interact (if at all)?   

Posted October 29, 2018| Permalink