Data Center Journal

VOLUME 56 | AUGUST 2018

Issue link: https://cp.revolio.com/i/1012043

Contents of this Issue

Navigation

Page 10 of 20

8 | THE DATA CENTER JOURNAL www.datacenterjournal.com A s we head into another hurricane season, the threat of power issues, damaged buildings, IT outages and data loss is looming heavily. Last year, Hurricane Maria devastated Puerto Rico and le more than 11% of the island without power for more than six months, affecting families, homes and businesses. e hurricane season of 2017 was among the worst in U.S. history, with 17 named storms. But according to Global Weather Oscillations, the 2018 hur- ricane season will be as destructive, or more so, than last year's. To protect businesses and data centers from down- time, it's important to prepare for any potential impact from storms, regardless of whether one hits. Here are some tips to ensure that you're ready to recover from any disaster and keep your organization uninterrupted. EVALUATING YOUR IT-RESILIENCE PROVIDER Hurricane season is prime time for an onslaught of outages. If this fact can teach organizations anything, it's that a stable and reliable disaster-recovery (DR) solution is imperative. Beyond catastrophic natural disasters such as hurricanes, other scenarios—such as simple yet common power outages—manmade problems stemming from human error or criminal activities, such as ransomware attacks, can lead to downtime. No matter the cause, however, experts across the board continue to advise the same thing: organizations need a full DR plan to ensure resilience and minimize downtime. Busi- ness leaders must take the time to evaluate their current DR and resilience technology to ensure they're up to date and able to restore all the company's data. If not, researching or upgrading to a top-of-the-line, trusted IT-resilience provider may be the answer. Once the technology backbone is in place, recognizing organizational trigger points becomes critical to creating an overarching DR plan. Each organization should perform risk and business-impact analysis to this end. e analysis will provide insight into the impact each system or application will have on the company if it goes offline and at what stage initiating recovery processes makes sense. DR PLANNING AND AUTHORITY MANAGEMENT What oen becomes apparent in an outage is an orga- nization's reliance on just one or two IT personnel to guide it through the DR plan. is approach is highly vulnerable. If the resident experts are unavailable owing to illness or vacation, or they're otherwise unreachable in a crisis, the po- tential for prolonging and exacerbating the impact increases considerably. e way to avoid this dependence is thorough docu- mentation that gives all, or at least most, of the IT team the ability to understand all DR procedures to ensure minimal interruptions. e size and span of those involved with the DR plan will vary widely depending on the company's size and verti- cal market. For example, most Fortune 100 companies will have staff dedicated solely to business-continuity and DR planning. Yet no matter its size, a company must document the names, titles and contact information for all employees with disaster-management authority. TESTING YOUR DR PLAN THROUGHOUT THE YEAR Testing the DR plan still isn't a priority for all com- panies, even though it's the most critical part of the entire business-continuity plan. Ideally, a company should test its plan once per quarter, but if this frequency is unviable, twice per year is the absolute minimum. In the case of those who oen find themselves in the path of hurricanes, testing should be as oen as once per month—and specifically in the run-up to hurricane season. In addition, any time the company's infrastructure adds or changes a major vendor or supplier, DR testing is important. Testing the DR plan still isn't a priority for all companies, even though it's the most critical part of the entire business-continuity plan. Ideally, a company should test its plan once per quarter, but if this frequency is unviable, twice per year is the absolute minimum. In the case of those who often find themselves in the path of hurricanes, testing should be as often as once per month—and specifically in the run-up to hurricane season.

Articles in this issue

Links on this page

Archives of this issue

view archives of Data Center Journal - VOLUME 56 | AUGUST 2018