Colocation Selection: Analyzing the Workload
November 20th, 2012 By: Bill Kleyman
In selecting the right type of data center colocation, administrators must thoroughly plan out their deployment and strategies. This means involving more than just facilities teams in the planning stages. The process to select a good data center has to involve not only the physical elements of the data center, but the workload to be delivered as well. This is the third in a series on Colocation Selection and Best Practices. Below is an abridged version of Setion 3 of the Data Center Knowledge Guide to Colocation Selection.
There are several key considerations around the type of data or applications an organization is trying to deliver via the data center. Are you working with web applications? Or, are you delivering virtual desktops to users across the nation? In making these preparations, engineers and facilities personnel should consider the following important decision points.
Cloud and Big Data
Cloud computing will continue to evolve and change how organizations distribute their data. The idea behind cloud computing is that information becomes distributed between multiple locations and can then be delivered more creatively.
IT consumerization and BYOD (Bring Your Own Device) have driven the need for cloud computing as today’s IT environment continues to evolve. The current business landscape has created what is known as a “data-on-demand” generation where information is needed immediately on any device, anywhere and at any time. Because of this culture, it’s important to evaluate the following two components when selecting a colocation provider:
- Data center distribution.
- Big data.
Mission Critical vs. Good Enough
Take the time to understand the workloads, applications, and data instances which will be running within your data center. This means classifying the workloads based on their priority. In those situations, it’s important to conduct a Business Impact Analysis (BIA) to truly understand the value of various data and infrastructure components. The results of a BIA are very descriptive and indicate which systems should run in high-availability
Tier III data centers. Furthermore, it will help determine which systems must have higher uptime requirements versus those with a lower priority.
Network Bandwidth and Latency
With the increase of traffic moving through the Internet, there is a greater demand for more bandwidth and less latency. As discussed earlier, it’s important to have your data reside closer to your users as well as the applications or workloads which are being accessed. Where data may have not fluctuated too much in the past, current demands are much different.
Balancing the Workload, Continuity and Disaster Recovery
Selecting a colocation provider goes well beyond just choosing their internal features and offerings. Companies looking to move to a provider platform must know what they are deploying, the continuity metrics of their infrastructure and incorporate disaster recovery into their planning.
Various technologies can affect how well a data center performs. The distance the data has to travel and the amount of bandwidth provided by a colocation provider can mean the difference between a great user experience and a failed colocation deployment. Cloud computing has created a greater dependency on WAN technologies and virtualization has enabled significantly more powerful servers and more dense storage. With these new technologies come new considerations around how this type of data is being stored and delivered. When selecting the right colocation provider, make sure that their infrastructure is capable of growing with the needs of your organization.
To get more insight on how to analyize the workload and other colocation selection best practices download the complete Data Center Knowledge Guide to Colocation Selection.