Experience Providing Infrastructure for Many Pays Off for Every New Client
A data center for a mid-sized manufacturing company may be serving 5,000 users of IT services, whereas an online retailer may have a half-million to a million unique visitors in a year.
When you’re a cloud services (CSP) provider for mid-market companies, you may have 3,000 customers, plus all your customers’ customers depending on you to do your job well. Many industries and business types will be represented among the CSP’s customer base, each unique in its own way across a broad swath of variables. The CSP must account for and be accountable to all of them, satisfying their business objectives and meeting their expectations as if each one was the only cloud tenant.
The scope of this responsibility is never far from the minds of Peak 10’s solution architects and cloud infrastructure engineers or, for that matter, of everyone who works at our 25 data center locations in 10 U.S. major markets. Among the biggest challenge is keeping the promise of having state-of-the-art systems and software ready and available for customers when they want them.
Infrastructure and systems maintenance programs are continuous, while technology refresh, capacity expansions and business investments go on continually. While a customer is concerned only with its own operations, as it should be, the CSP must consider and evaluate what the impact might be of any change on all customers.
At Peak 10, there are four notable checks and balances among many operational routines that help us maintain an even keel.
Top-Flight Technology Partners: This is clearly a case of one plus one equaling three. Things do go wrong, regardless of whose equipment is involved. It’s how problems get resolved that make the difference. For example, during a routine firmware upgrade at one of our data centers, an infrastructure engineer realized something was not right. One of two main modules in the infrastructure was dead. Over the course of many days, working in constant contact with Cisco engineers and orchestrating a planned outage, this issue was resolved without any effect on our customers. Even though they never knew anything was amiss, we notified customers of the situation anyway.
MOPping Up: Maintenance operations protocols (MOP) define in detail the procedures to be followed for expected and unexpected events. As Peak 10’s infrastructure grows, new systems come online and new services are created, MOPs are essential for tasks such as loading software, problem diagnosis or orchestrating planned outages in the event of hardware failure as mentioned above. It’s critical that we have and can demonstrate specific documented procedures for handling particular situations.
Change Advisory Board: Any activity, project or process change with potential customer impact must go through a Peak 10 Change Advisory Board. Controlling change is critical because it can have unanticipated consequences across the organization. Change proposals must be vetted and documented and, in some cases, replicated across multiple locations. Again, expanding infrastructure across multiple geographies make change management increasingly critical to keeping our customers happy.
Culture: From the first day of new-hire orientation, the companywide commitment to customer focus is instilled as a core operating principal, whether an employee’s role involves direct customer contact or not. It’s a responsibility that everyone understands and takes seriously.