System Design · ByteDance · Medium
Design a data-warehouse platform by deciding what role each candidate technology should play, given a mixed workload that includes batch transformation, interactive exploration, and scheduled reporting. Candidates under consideration are Hive, Spark, the storage and execution layers of the Hadoop ecosystem, and SQL Server. Describe which duties each can realistically shoulder, which pairings make sense together, and how workload shape, data scale, latency expectations,…
Checking your access…