• Hadoop, NoSQL & NewSQL
  • Experfy Editor
  • APR 23, 2014

Big Data Database – To SQL or NoSQL?

Wikibon study on the big data market conducted around the first quarter of 2013 attempted to divide the market—based on the database market shares it holds. The study report forecasts the size and projected revenue of SQL and NoSQL database markets as related to big data.

Since 2011, both SQL and NoSQL database technologies have made strong contributions to the growth of the big data market. Currently it may appear that the enterprises are broadly divided into SQL and NoSQL factions, but the fact remains that each of these database technologies has its relative merits and demerits. They both have particular characteristics that fit specific types of big data applications. While SQL databases are known for data consistency, a richer programming environment, and easier testing and maintenance; NoSQL is better for scalable solutions and cost-friendly, big data applications.

The challenge  most enterprises face today is selecting the relevant database technology for specific business application scenarios. Both SQL and NoSQL databases have value within their respective fit areas in both transactional and analytics big data.

 

Market forecasts for SQL databases for big data 

A relational database scripted by IBM, SQL upholds a host of industry standards, which are somewhat ambiguous! As SQL database vendors do not follow the same standards, vendor lock-in is a common problem.  The leading SQL software vendors include IBM, Oracle and Microsoft, who offer very costly products with the capability to handle highly complex queries and transactions.

The Wikibon study shows that Oracle software is so costly that the combined cost of servers and the infrastructural software constitutes only 10% of the total cost of big data database implementation. This cost constraint creates a serious challenge to traditional SQL in big data systems. Moreover, the technical limitation of SQL database technology has been discussed in great detail in the Wikibon study on the big data market.

 

Market forecasts for NoSQL databases for big data 

The birth and evolution of the NoSQL databases happened as an answer to challenges posed by big data. The large online businesses like Facebook and Google discovered NoSQL by accident—when they could not handle huge volumes of data with the conventional SQL databases. The drawbacks of the NoSQL systems are lack of schema or data consistency, difficulty in testing and maintaining, and absence of a structured query language.

However, NoSQL databases are apt for processing very high volumes data, while ensuring performance and speed of execution. The NoSQL database market for big data has maintained high growth rate since 2011.

  Source: Big Data Database Revenue and Market Forecast 2012-2017

Source: Big Data Database Revenue and Market Forecast 2012-2017

 

 Top big data database vendors in ranked order

The top 6 vendors endorse SQL databases:

  1. IBM commands a sizeable 18% of the big data database revenues—generating business of $215 million.
  2. SAP also enjoys 18% of the big data database revenues at $190 million from its SQL products.
  3. HP with Vertica commands 12% of the big data database revenues, at $150 million from SQL.
  4. HP with Terradata —generating $122 million from its wide installation base.
  5. EMC generates $105 million, which amount to 9% of the big data database revenues.
  6. MarkLogic, though trailing in the sixth position, is the leading NoSQL vendor. It commands a $43 million market amounting to 4% of the big data database market.
Boston city bkg

Made in Boston @

The Harvard Innovation Lab

350

Matching Providers

Matching providers 2
comments powered by Disqus.