Philip Russom, research director for data management (DM) with TDWI Research, is a veteran myth-buster. For a few years now, Russom has been battling Hadoop-related myths and misconceptions. Some are especially hard to bust, he says. Call them “zombie” myths or misconceptions: the ones that refuse to die.
The biggest “myth-conception,” says Russom, concerns what people are actually doing with Hadoop. From the beginning, it’s been closely associated with MapReduce, its massively parallel general-purpose execution engine. The overwhelming majority of adopters are using HDFS—and just HDFS, along with the Hadoop kernel. In most use cases, they’re using Hadoop’s core file system as a versatile and practical means of storing stuff.
This doesn’t mean they’re using Hadoop as a “database,” however.
READ MORE: Download this Hot Topic.
Sponsored By Teradata
Individual, Student, & Team memberships available.