Data warehouse appliances are catching on because they're simpler to use and administer than the traditional data warehouse. They're also easier to scale. But that doesn't mean you get a pass on the integration and migration headaches, according to Evan Levy, partner and co-founder of Baseline Consulting.
In a recent two-part series, Levy examines this trend to switch from a traditional relational database to data warehouse appliances. In the first part, he provides historical context for data warehouse appliances and looks at what's driving their adoption. I suggest you read the whole post, but basically the drivers are scalability and simpler management of platform and system resources-plus, as with all things tech, a bit of marketing hyperbole:
"The glossy sales message of data warehouse appliance vendors comes down' to something like: 'We've reduced the complexity of running a data warehouse. Just install our appliance like a toaster, and watch it go!'"
It's a message that has fooled some early adopters into believing they can "simply fork-lift their existing data warehouse structures onto their new appliance," Levy says. This misconception can actually impact the appliances' longevity and performance or price rationale, he adds.
How do you avoid that mistake? Understand that you still have to think through the data architecture issues. To a large extent, that's going to mean reviewing the ETL architecture and processes issue, as Levy explains in the second part of his series:
"Each appliance has its own way of handling the intensive resource requirements of data loading-in much the same way that each incumbent database product addresses these same situations. If you've justified an appliance through the benefits of consolidating multiple data marts (that contain duplicate data), it only makes sense to consolidate and integrate the ETL processes to prevent processing duplication and waste."
Yes, these pieces are a bit more technical than a business manager would care about, but it's an excellent read for IT leaders and middle-managers, and a must read if you're considering a data warehousing appliance. After reading Levy's posts, it's obvious that this would be a project for which it's easy to underestimate, understaff and under-train.
You might also want to read our Arthur Cole's post "Can Warehouse Appliances Really Get the Job Done?" Cole also contributes to Data Center Edge, a part of our IT Business Edge network and a useful resource for all things pertaining to data centers, including articles on hardware, appliances, virtualization and data warehouse.