The emergence of data warehouse appliances has broadened the potential uses of business intelligence (BI) and analytics within many organizations and is enabling companies to take advantage of data warehousing technologies that otherwise might have been beyond their reach. But limitations exist, and depending on an organization's goals and available resources, appliances may or may not be the right option.
Traditional data warehouse systems and appliances that bundle together data warehouse hardware and software in a single package provide many of the same potential benefits to companies. Whichever approach they choose, organizations use data warehouses to consolidate data from multiple source systems, manage data quality and integration processes, and support BI and analytics capabilities that enable business users to gain insight from the data.
In many cases, when businesses look at implementing a data warehouse appliance, they're seeking to expand an existing warehousing infrastructure with the goal of boosting performance, adding more storage capacity or enabling new types of analytics. Other organizations look to data warehouse appliances as a gateway to BI and analytics due to their perceived ease of implementation in comparison with traditional data warehousing.
The two technologies are similar in structure, but the traditional approach involves separate data warehouse hardware and software with in-house developers responsible for the design and development of the data warehouse architecture. With appliances, vendors provide a server (or set of servers) with optimized data warehouse software and the warehousing structure already in place.
When looking at which approach to choose, smaller businesses or companies looking to provide analytics capabilities to a specific department or business unit are flocking to appliances because of their deployment advantages. And with “big data” analytics becoming a priority for many organizations, appliances are being used to consolidate large amounts of information as an extension of traditional data warehouses. As both vendors and user organizations focus more on appliances, the packaged systems likely will play an even bigger part in data warehousing and BI initiatives going forward.
Understanding the potential benefits of data warehouse appliances can help organizations identify whether they're the right fit to address business requirements and issues. For example, for businesses that are trying to solve targeted issues such as gaining better insights into customer data for marketing uses or looking at ways to improve customer satisfaction, the ability to quickly deploy a data warehouse appliance and manage it separately from an existing data warehouse could provide value that might be more difficult to achieve by expanding the current infrastructure to support such uses.
Taking the full measure of data warehouse appliances
But even though appliances can address many business needs, there are situations in which organizations might choose not to adopt them in lieu of a traditional data warehouse. For organizations with mature data warehousing and BI environments, looking toward appliances as an additional component to use in expanding their infrastructures might not be feasible for financial or technical reasons. And not all organizations are poised to take advantage of more advanced data warehousing technology, such as columnar databases and in-memory analytics, that many appliances support.
The truth is that in addition to their potential benefits, data warehouse appliances do have some limitations to keep in mind when considering whether or not they make sense for an organization. For example, although some appliance vendors provide a variety of server choices for hosting their data warehouse software, most limit their offerings to one or two hardware options.
Also, despite the perceived ease of use in comparison with building and managing traditional data warehouses, organizations might require new skill sets to install and maintain appliances. That could mean hiring new workers, which might put the long-term cost of an appliance deployment out of reach despite the lower pricing structure for the technology itself. It's also important to assess the ability to expand the available storage space in appliances and issues that might affect overall performance as workloads increase. (The same can be said of traditional data warehouse systems, of course.)
With so much attention being focused on data warehouse appliances these days, it's easy to get caught up in all of the hype surrounding them. But taking full stock of the potential pluses and minuses of appliances is essential in helping organizations to decide whether the data warehouse hardware and software bundles are the right choice for them.
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.