If you’re pulling data — particularly large volumes of disparate data — together to make analytics and reports, you’ll need to store it somewhere. As a core component of data-driven business intelligence, a Data Warehouse (DW), combined with an ETL (extract, transform, load) platform, integrates data from different, typically transactional, sources (like Marketo, CRM, Salesforce, inventory systems, and even web analytics, to name a few). A DW often stores both current and historical data, and is always aimed at getting overviews and insights only possible from a single, canonical data store.

When choosing a DW solution for the first time, the very first consideration is typically one between an on-prem DW or a cloud-based one. And while a lot of folks brand new to the domain go straight to the cloud these days, there are still many reasons why you might want to choose an on-prem solution. Are you comfortable with your data in the cloud, or would you prefer to keep it on-premise? Are there compliance requirements that you may find easier to ensure with an on-prem solution? Are you running legacy systems that do not integrate well with cloud offerings? Are your data volumes high enough (read: lots of connected devices) to justify the scale of the cloud, or small enough (read: mostly local, transactional and/or operational) such that you can comfortably keep your data in-house? Do you need the support that a cloud offering provides? Note that most on-prem solutions these days do offer some capacity to partition and scale, but also note that you (or your team) will be the ones doing it.

Source: DZone