IT-SuccessWin Technology

"What are the best books for learning Python?" in   IT Courses/Programming Languages/Python

I have found the book by Mark Lutz to be a very good reference. If you want to learn scientific Python, try the online free books below: Scipy Lecture Notes Table of Contents of Quantitative Economics Python for Econometrics

"Which Python should I learn, Python 2.7 or 3.x?" in   IT Courses/Programming Languages/Python

If you're learning Python irrespective of a specific purpose, go ahead and learn Python 3.2. As Harel Malka noted, once you have the fundamentals of the language down, you can learn the differences between 2.x and 3.x as needed. The hard part is learning the language as such. If you need Django or similar modules or frameworks that do not (yet) have 3.x support, though, then learn using 2.7. (It was a while before I moved to 3.x because I needed Numpy and Scipy support, for example.) When it comes time to move to 3.x, you'll have the fundamentals down and the differences will largely be smallish annoyances rather than sizable stumbling blocks.

"What is the best for Java developer in between AWS and Hadoop?" in   IT Courses/Hadoop

IMHO learning Hadoop or AWS is like learning SQL with Java. These are additional skills with Java work. AWS: You can start with AWS if your organization has already selected AWS for their infrastructure. AWS is ever growing field. You can first familiarize yourself with the scope of AWS associate architect. This area touches a bit of everything. Hadoop: You can pick Hadoop if you see yourself working on large scale projects with huge data. Hadoop is just the beginning. Depending on your interest you can pick Hive, Scalia, Spark to grow your knowledge of Hadoop ecosystem. I think both AWS and Hadoop have great potential. You will not regret learning any of these.

"What are the advantages of using ORC file storage format in Apache Hive?" in   IT Courses/Hadoop

Speed. Reading data from well partitioned, indexed, aggregated file will return very fast results

"What is the advantage of using Apache Spark over Elasticsearch for analysing data?" in   IT Courses/Hadoop

Those are two entirely different technologies and serve different purposes and used for different use-cases. While Elasticsearch is a full-text search engine, Spark is a distributed in-memory computing framework for doing advanced analytics workloads such as interactive or ad-hoc queries, batch and streaming ETL, and machine and deep learning applications. To put it simply one is to distribute your compute operations to all the workers or nodes in the Spark clusters, the other is to do your text search on a large corpus of text stored as books or indexed as text files. This not to say that they cannot be used together for certain use-cases. For instance, you may be doing some semantic analysis using Spark’s MLlib library, and consult Elasticsearch for some ranking or relevancy or search frequency of certain words in the text. Both serve different purposes, yet can work together. Cheers Jules