Experience with Hadoop by a novice user.
May 26, 2016
Experience with Hadoop by a novice user.
Score 7 out of 10
Vetted Review
Verified User
Modules Used
- Hadoop MapReduce
Overall Satisfaction with Hadoop
Hadoop is not used as a norm in my organization. I just use it personally to complete my job faster. It is implemented in the research computing cluster to be used by faculty and students. It completes jobs faster by parallelizing the tasks using MapReduce framework. This gives me considerable speed in the tasks I perform.
- Provides a reliable distributed storage to store and retrieve data. I am able to store data without having to worry that a node failing might cause the loss of data.
- Parallelizes the task with MapReduce and helps complete the task faster. The ease of use of MapReduce makes it possible to write code in a simple way to make it run on different slaves in the cluster.
- With the massive user base, it is not hard to find documentation or help relating to any problem in the area. Therefore, I rarely had any instances where I had to look for a solution for a really long time.
- I would have hoped for a simpler interface if possible, so that the initial effort that had to be spent would have been much less. I often see others who are starting to use hadoop are finding it hard to learn.
- I'm not sure if it is a problem with the organization and the modules they provide, but sometimes I wish there were more modules available to be used.
- I did not have to invest anything in particular. I was able to complete tasks much faster with Hadoop, which helped me save a lot of time.
- When I started using Hadoop, I was trying to finish a project that was taking too long to execute. But since I was not so experienced with Hadoop, I took a long time to use it. This made me miss the deadline.
- MapReduce and Amazon Elastic MapReduce
Hadoop was a cheaper alternative to Amazon. Since I had to pay for every minute I use with Amazon, I had to make sure multiple times that the code was good enough before I purchased with Amazon. But since Hadoop was available on the cluster, I had the opportunity to code on the way.