Are data warehouses of a bygone era?
We've been talking to many customers and what seems to be cropping up time and again is a discussion about their data warehouse. In almost all these discussions two things seem to stand out:
- Their data warehouse is consuming a vast amount of storage and/or,
- their data warehouse is underpowered when there is a surge in demand
Of course, the data it contains are important to the business, but how does the DW scale? In traditional DWs, compute and storage are tightly coupled. More storage comes by default with more compute. More compute generally means more storage. Either way, scaling up can be relatively pain-free, but scaling back is usually not.
Added to this, scaling up is costly. More hardware, more disk, and if you're not in the cloud, then possibly the most painful part is procurement.
Splitting storage and compute can help solve both these problems. Using a data lake, processing with Apache Spark, and using Amazon EMR, your dynamic data warehouses can scale to meet your business demands while driving costs in the right direction - and there are other benefits too.
Read more about how Cloud-Fundis have helped customers move closer to that goal.