Myths on big data Avoiding bad Hadoop and Cloud Analytics decisions
Myths on big data Avoiding bad Hadoop and Cloud Analytics decisionsYou had better read these myths about big data analytics to avoid bad Hadoop and Cloud Analytics decisions.
It may be true that Hadoop is an open source legend that was built by software heroes. Nevertheless, legends can sometimes be surrounded by unfounded beliefs, these myths can lead IT teams down a path with rose-colored glasses. Here are those myths which are worth having a look at.
1. Big Data is purely about volume.
In addition to volume, several industry monitors have also touted variety, variability, velocity and value. The point is that data is not just increasing, it is moving further towards real-time analysis, coming from structured as well as unstructured sources, and being utilized to try and make better decisions. As a result, analyzing a big volume of data is not the only way to reach the value.
2. Traditional SQL does not work with Hadoop.
When Facebook, Twitter, Yahoo and others bet on Hadoop, they were also clear that HDFS and Map Reduce were restricted in terms of their capability to deal with expressive queries through a language such as SQL. This is how Hive and Pig were ultimately hatched and many companies as well as projects are providing methods to solve the compatibility of Hadoop and SQL.
3. Hadoop is the only new IT data platform.
There have been many investments in the IT portfolio, and the mainframe is an example of one which probably should evolve along with ERP, CRM and SCM. As the mainframe is not being eliminated by companies, it obviously needs a new strategy to grow new legs as well expand on the value of its current investment. According to many customers who run into problems with mainframe speed, scale or cost, there are increasingly ways to evolve the big iron data platform and consequently get more benefits out of it. For instance, in-memory big data grids such as vFabric SQLFire can be implanted or use distributed caching approaches in order to deal with such issues as high-speed ingest from queues and real-time analytical reporting.
4. Hadoop does not make financial sense to virtualize.
Hadoop is typically supposed to run on a bank of commodity servers; thus, it might be concluded that adding a virtualization layer adds extra cost but no extra value at all. There is a flaw in this perspective that you are not considering the fact data and data analysis are both dynamic. In order to become an organization that motivates the power of Hadoop to develop, innovate as well create efficiencies, you are going to vary the sources of data, the speed of analysis and more.
What is more, virtualized infrastructure still lowers the physical hardware footprint so as to bring CAPEX in line with pure commodity hardware, and OPEX is also reduced through automation and higher utilization of shared infrastructure.
5. Hadoop does not work on SAN or NAS.
Although operates on local disks, it can also work well in a shared SAN environment with variable cost and performance characteristics.
Download and use it now: Myths on big data Avoiding bad Hadoop and Cloud Analytics decisions free download
Add a review
Tell us your experience with Myths on big data Avoiding bad Hadoop and Cloud Analytics decisions
Related programsOur Recommendations
If you think that creating a video with your photos is too complicated, think again and take a lo...Download
Diagram Designer is an easy-to-use tool to create all kinds of diagrams, charts and other schematic designs that you can use for different purposes: school lessons, work presentations, design, etc. Diagram Designer's interface is very intuitive. Simply drDownload
Easy software for managing digital photos/short videos with web connectivityDownload
a suite of freeware and shareware tools for automating work of rendering and animating moleculesDownload
Ashampoo Snap is a feature rich screen capturing tool, for taking screenshots and screencasts.<... Download