So, with this blog, the end of my Cisco Domain Ten(SM) journey approaches. Starting back in December, I’ve now covered each of the ten domains in our Cisco Services’ framework for data center transformation. A few weeks back, I discussed some of the challenges you may face in Domain 9: Security and Compliance, and it’s now time to discuss Domain Ten: Process and Governance, and the overarching challenges of operations management. This is a hugely important topic so in this case I’ve split my blog into 2 parts, with this part (a) focused on your challenges, and the subsequent part (b) aimed at outlining how we in Cisco Services can help you address these operations management challenges.
The Cisco Domain Ten framework applies as much to service provider data center transformation as it does to business (enterprise) data center transformation. In my Cisco Domain Ten series, I’ve often focused on enterprise challenges, so for this blog, I’ll focus on the operational challenges I’ve observed at first hand in the service provider environment. That said, most if not all points I raise in this article will be equally applicable to enterprise/business and public sector data center operations.
When discussing technology introduction, it’s often too easy to focus myopically on the technology itself and capital expenditure. What about the people and process issues associated with introducing new technologies and concepts into your data center? What about operational expenses? Great questions and this is what I will expand upon here.
As with my previous blogs, I will consider cloud computing as my “use case”, say for example you are introducing new converged infrastructure to roll out cloud services to your customers. I’ll discuss some of the “Process and Governance” questions that may be on your mind, and be open on some of the problems that may be prevalent in your operations teams. Note that, as with previous blogs, my list here is not an exhaustive list of operational challenges, as I’m sure you will appreciate, so please forgive me if I omit your major pain point!
- Firstly, it’s not just about technology. There are people (organizational) and process challenges that if you don’t solve properly, you may never meet your application SLAs, nor achieve return on your CapEx investment. For example, are your operations support teams aligned to deal with converged infrastructure? Or will those network, compute and storage team silos still in place hold you back from achieving those SLAs and trouble ticket response times? Are the tools, processes, KPIs and SLAs still valid and appropriate for the new environment?
- Are your existing operational processes a bottleneck? For example do you have (say) a 4 hour SLA … and find problems being escalated to “level 2” support at 3 hours and 56 minutes? (giving your poor, swamped level 2 support team no chance of meeting the SLA?)
- How can you adopt and exploit the best practices in ITIL?
- Do you have the right skillsets and automation tools in each of your support levels? (Most if not all SPs I’ve worked with have multi-layer support organizations, where level 1 have the initial look at the problem before passing on if they can’t solve it). Have you standardised around tools and scripts to help you troubleshoot, so that the information about a trouble ticket is passed from level 1 to level 2 support? Or – as I found – do your level 2 technicians basically repeat everything that level 1 support should have done, because they don’t trust the first team’s rigour?
- I mentioned the “C” word above – “CapEx” or Capital Expenditure. Many businesses focus primarily on the Capex , that is, the cost of product acquisition, when planning say a new cloud service rollout, however this is not usually where the majority of costs lie. I worked with service providers for many years and it wasn’t uncommon to see data showing that CapEx was less than 30% of overall costs. So ignore the operations management challenges and total cost of ownership at your peril.
- Once you have the new equipment in place, are you sure you know all about the management features that will help you be more responsive. Do you know which SNMP traps from the new equipment to pay attention to? Do you know which to ignore or filter out? Does your team? Is such action defined in your operational processes and Standard Operating Procedures (SOPs), so that when you on-board new staff, they also know what to do? Does your fault management system deal with all the new alarm conditions from the new converged infrastructure and cloud equipment? Do your service assurance tools give you the visibility into the cloud?
These are all very relevant concerns and questions that outline the challenges of “Domain Ten: Process and Governance”. Net-net is that if you don’t address challenges in this domain, you could really nullify your investment in domains 1 – 9! Operations touches all the other 9 domains -- there is an operations element to anything that you want to put into production.
Next week I’ll finish off my Cisco Domain Ten series and tell you about how our new Cisco Data Center Services for Operations Enablement help you address the types of challenges I’ve outlined above, and how they provide solutions for this “Domain Ten: Process and Governance”.