Choosing the right data warehouse technology is no easy task. Whether that's traditional data warehouse software...
or a data warehouse appliance, evaluating and selecting the product offering that best meets the business requirements of an organization involves a variety of steps and considerations – and some challenges.
For example, access to information about data warehouse appliances is abundant, through media articles, vendor white papers, product demos and the like. Sometimes, however, too much information can create an inability to identify how products really differentiate from one another and which one best addresses the business needs – and pains – that an organization is facing.
Identifying product differentiators is essential as part of the process of selecting data warehouse appliances, especially with vendors in general promising faster query response times, larger storage capacities and better support for advanced analytics. The reality is that with ongoing technology advancements, many vendors do offer similar basic levels of functionality, including the ability to customize their offerings to suit the requirements of individual customers. The true differentiators may result from support for things such as in-memory analytics, columnar databases and, most importantly, value-added services.
As the appliance market matures and more of the independent vendors are acquired by larger companies with broader data warehousing and business intelligence (BI) product offerings, much of the value of different technologies will stem from vendor services and the overall customer experience. Technical support, flexible licensing policies, data integration capabilities and the ability to influence future development and feature expansion could be the deciding factors that make one product more worthwhile than the others in a data warehouse appliance comparison, irrespective of initial pricing.
Some key areas to look at during an appliance evaluation and selection process include the following:
Your current IT environment and internal standards. Taking a close look at what already is in place internally can help focus the selection process on the data warehouse appliance vendors that best fit your current IT infrastructure. For example, many large organizations have existing hardware and software standards. Integration options and the ability to add new hardware platforms may be limited based on these standards, leading to a more defined approach to choosing an appliance.
Business requirements and project scope. By identifying the key goals of using a data warehouse appliance, organizations can hone in on the type of device that's required. Whether appliances will be used primarily for advanced analytics applications or to store large amounts of data for basic BI querying should drive decisions on the type of database that's used. For example, columnar databases are known for their analytics performance. Other considerations to take into account include the number of source systems that will feed data to an appliance and storage requirements.
Required skill sets and IT resources. One of the reasons why appliances are becoming more mainstream is that many businesses want access to terabytes of storage, advanced analytics capabilities and the benefits of data warehousing without having to build data warehouse systems in-house. Although appliances require fine-tuning as part of the deployment process, in general they hold out the promise of easier and faster implementations compared with traditional approaches. But even so, internal resources are required to implement and maintain appliances – depending on the product that is chosen, one to three IT staffers might be needed on an ongoing basis. In addition, specific skill sets could be required. That means organizations might have to limit their choices to match the skills they already have or broaden their internal capabilities through training or by hiring additional staff.
Future growth. With data warehouse appliances, much of the forward-looking aspect of selecting products involves the ability to anticipate data growth over time, whether that's through the addition of new data sources, business expansion or the need to incorporate more historical data. Organizations need to be aware of the differences between appliances: Some vendors support varying storage levels, while others require customers to add more hardware or upgrade to new appliances. With traditional data warehouses, businesses might have more flexibility in relation to how they choose to grow their data infrastructure. Companies that use appliances might have to tie growth plans to what their chosen vendor offers and hope that it meets future business requirements.
Overall, the appliance market has unique requirements that organizations need to consider when evaluating the bundled packages of data warehouse hardware and software. Each organization will differ in relation to the benefits and challenges associated with their search for the right data warehouse appliance. In the end, identifying key business goals and tying them to vendor sweet spots will help companies make the right appliance choice.
Lyndsay Wise is president and founder of WiseAnalytics, an independent analyst firm and consultancy based in Toronto that focuses on business intelligence and dashboards for small and midsized organizations.