|
7 Steps to Mission-Critical IT Services
CIOs today are being called upon to interpret business strategy and priorities, then add value through information technology initiatives. At a time when “critical to the business” translates to “mission-critical IT,” how do you ensure end-to-end availability and reliability of the IT resources that enable your company’s essential business processes?
This e-book offers seven steps for achieving continuous service availability to support key business processes. We’ll also examine the financial and other impacts posed by business service downtime. This series comes to you from Stratus Technologies, a leader in mission-critical availability for more than 25 years.
Step 1: Don’t assume. Assess.
At the same time expectations for service delivery are rising, CIOs are taking steps to bring IT into closer alignment with the business. And the more intrinsic IT applications and services become to business processes, the more they become mission-critical.
These increasing demands mean it’s probably time to reassess your mission-critical environment against the job you are being asked to do. For example, tougher mandates for information security and regulatory compliance require complete,auditable data. Not to mention that more users, more customers and more partners are probably counting on your IT services. If so, every hour of downtime will cost your company more.
Maximizing the end-to-end availability of your mission-critical environment involves more than technology, of course. Procedures, policies and how people carry them out every day are vital. The same goes for security measures. It is important to understand the state of these elements and the dependencies between them. An up-to-date audit of your IT environment provides a baseline for staying in step with business needs. Even when you believe your IT infrastructure is delivering optimal availability, a review can reveal new or overlooked information. Perhaps incremental changes such as software patches have introduced risk factors that have escaped notice. An assessment can also show whether previous efforts were as successful as they could be.
Step 2: Get the metrics right.
After taking stock of your current mission-critical IT environment, make sure you are measuring the right things.
When response time is of the essence but throughput is slow, users won’t care that you can prove the application, the servers, and the network are all up and running. When customers go to the ATM and can’t withdraw cash, neither they nor your VP of self-service banking will be impressed by everything that is working. They only see a problem. That’s why thinking like “the IT guy” is no longer enough these days. The business drives what was once directed and measured primarily in terms of technology.
Looking to ITIL disciplines Many companies look to the best practices known as ITIL®, the Information Technology Infrastructure Library, as a starting point. Our scope here can’t do justice to the topic of ITIL. But we would like to borrow a concept: Understanding the business needs of your organization is indispensable. That knowledge provides a base for understanding what continuous service availability means in terms of your business processes.
Take the vital step of communicating with your users to make certain you understand their requirements. When your goals involve existing IT services rather than new ones, you will want to know how expectations have changed over time.
More than just users’ concerns go unattended when you measure the wrong things. Your staff will be monitoring and reporting on metrics that may not have much relevance — time they could put to more productive use. Staff attention and IT policies may also be misdirected.
How many nines?
The gold standard of five nines, or 99.999%, availability is a familiar one. Designing individual components that can measure up to five nines — say, a server — is entirely possible. Now consider the chain of service delivery from end to end, and it is reasonable to expect the typical IT shop sees less uptime than that. You can’t count on all the links in the chain playing well, and you can’t count on them playing well together.
Each additional nine of end-to-end uptime for an IT service comes at a cost. Not to mention that it’s challenging to maintain. On the other hand, downtime has consequences that range from disgruntled users, to gaps in regulatory compliance, to financial losses.
|
