As we discussed, a single tool can't do everything. You need to use the right tool for the right job, and the data lake enables you to build a highly configurable big data architecture to meet your specific needs. Business problems are far too broad, deep, and complex for one tool to solve everything, and this is especially true in the big data and analytics space.

For example, hot data will need to be stored and processed in-memory, so caches or in-memory databases like Redis or SAP Hana are appropriate. AWS offers the ElastiCache service, providing a managed Redis or Memcached environment. NoSQL databases are ideal when facing high velocity but small-sized records, for example, user-session information or IoT data. NoSQL databases are also useful for content management to store data catalogs.