Hadoop quick review
February 23, 2016

Hadoop quick review

Anonymous | TrustRadius Reviewer
Score 9 out of 10
Vetted Review
Verified User

Modules Used

  • Hadoop Common
  • Hadoop Distributed File System
  • Hadoop MapReduce
  • Hive, Pig, Flume, Spark etc. All the ecosystem, HUE

Overall Satisfaction with Hadoop

We have Hadoop pre-prod and prod clusters. Production clusters are comprised of 200 nodes. And we have realtime clusters as well. All the data will be moved to Hadoop. We use Hadoop to do machine learning and data warehousing.
  • Machine Learning Model, when SAS can not process 3 of years data. Hadoop is good tool to build the model.
  • Data warehousing is also another good use case. Using Teradata is expensive.
  • A lot of people are not from a programming background which makes Hue very important for end users when starting the Hadoop journey. Making Hue more user friendly and functional will be helpful for end users who don't much of a programming background.
Data is growing and grows fast. A relationship database can't hold this requirement any more. Real-time applications and distributed design are required for highly scalability and fault tolerance.