Big data brings big management reform

Big data started to emerge in 2011, and in 2012 it has skyrocketed, and one day it may change all aspects of data management in a dramatic way. Big data systems bring visible changes in managing and manipulating computer data, continuous abstraction, transformation and loading capabilities, operational BI, dynamic big data and cloud-based data warehousing.

However, as big data goes into 2013, there is no system technology that is more active than the NoSQL database and Hadoop framework and there seems to be room for further growth in both products. According to a 2012 report by MarketAnalysis.com, the Hadoop-MapReduce market alone is forecast to grow at a CAGR of 58% and reach $ 2.2 billion in 2018.

The emergence of NoSQL and Hadoop is mainly to deal with unstructured data, such as text data or web log. Like Apache Hadoop, these technologies are usually started from an open source and gradually become new commercial products.

Judith Hurwitz, president and CEO of Hurwitz and Associates, a Massachusetts-based company, believes big data architecture and massively parallel processing have dramatically changed the picture. "Before then, even before data really matters for the company, she said," people do not have the power to get huge amounts of data and do real-time analytics. "

She said: "Now, unrealistic things are becoming practical.This situation has brought the data out of the comfort zone."

SQL hit, about to fight back

As we can see on the site, from the beginning of 2012, a prediction of the predicament of mainstream relational databases emerged. Part of the prophecy has become a reality. After a series of struggles with products that may become alternatives in the years to come, SQL relational databases appear to be facing (or soon) the strongest competition regarding the filtering of large amounts of data throughout the enterprise.

The driving force behind this trend is the desire of businesses to get more unstructured data at a faster rate so that businesses can rely more on data-driven decision-making. The usual treatment is changing to accommodate the best new technology.

These moves from specific data management vendors in 2012 show the status of big data and Hadoop impact on relational data:

IBM is also continuing to create smaller data and analytics companies, albeit less than in 2011. Blue Giant's efforts from small improvements (such as NoSQL for DB2 10 graphics library and InfoSphere Warehouse 10) to a very large PureData system installations, the purpose is to get big data for the enterprise.

Oracle introduced big data devices early in the year. This release was released immediately following Oracle NoSQL Database 2.0, which automatically rebalances Oracle NoSQL Database 2.0. The new application programming interface handles large objects and has tighter integration with the Oracle database, allowing direct SQL queries of Oracle NoSQL database records.

Microsoft presented a preview of Hadoop support for Windows Azure and Windows Server; Teradata released its Aster big data analytics product; and Informatica released the big data edition of the PowerCenter suite, purportedly eliminating the need for Hadoop hand-coding and programming Mission into the Informatica development environment.

SQL may have only one or two counterattacks in 2012, but its aggressiveness in responding to market challenges is important. Companies that are more professional at non-mainstream NoSQL and Hadoop have updated their SQL credentials last year. A typical example is Hadoop's creation of Cloudera, which is looking to increase the level of collaboration between SQL and Impala (Impala is a Hadoop software product that supports standard SQL for interactive queries).