RDBMS: No Full Stop

Relational Database Management System
(RDBMS), which is primarily known as
high-performance, secure storage area, has now evolved into a highly elastic solution consisting of OLTP storage, analytics, data integration, warehousing and reporting.

Today, RDBMS goes far beyond providing mere storage and retrieval of data. It is seen as data warehouse appliances, which typically consist of database software combined with pre-configured hardware, OS, software and support services. Besides, RDBMS vendors have come up with several high availability and DR solutions like grid computing and clustering solutions. These lead to consolidation of multiple databases on a single platform and higher uptime for mission-critical systems. Moreover, new functionalities are being continuously added in the RDBMS. A key technology being incorporated into databases is enhanced business intelligence (BI) capability. This is a high-value add as BI can provide significant insights on making the businesses more efficient.

Kamal Vohra, Lead Analyst-Software and Services Research, IDC India, said, “RDBMS continued to be the most widely deployed database management system in India in 2010. Traditionally, large organizations have been the spenders on these tools. As the IT maturity level of mid-sized enterprises is beginning to increase, the use of RDBMS tools is increasing in this segment. With huge volumes of data flowing in their large enterprises, the need for an efficient and scalable database management system has come to the forefront. Demand for data searching and linking is increasing even at the organization level.”

A preferred platform

Going by the adoption levels amongst Indian enterprises, RDBMS enjoys a higher degree of preference as compared to other database management systems. RDBMS tools have been in extensive usage for the past many years, wherein data can be written faster onto a relational database; the structured query language (SQL) is simple, provides an easy mechanism to relate and find bundles of data, and the database is readily scalable.

“The moment end-user's productivity is enhanced, the solution becomes useful and well used. Take the example of BI today from Microsoft's SQL server. With capabilities like PowerPivot, which help integrate and analyze inputs from several data sources into more than three million rows of Excel, to ad hoc reporting and detailed statistical analysis directly from the SQL server warehouse, help end-users to enhance their decision making capabilities which in turn make them more productive,” pointed out Pallavi Kathuria, Director-Server and Tools Business, Microsoft India.

Nirav Shah, Principal Consultant, ADP India, further added, “The relational database management software continues its leadership in the market globally since several decades. The flat files and hierarchical databases are more suitable for small applications and have limited scalability. For any enterprise-level application the need for data integrity, concurrency, and the independence of logical data model from physical storage of data combined with performance is paramount.” Here, if we take a look at the current state of RDBMS deployment amongst the Indian Inc, we will find that the market leadership of RDBMS across the entire IT industry is echoed in India Inc too. According to market search data for FY '09, the growth enablers for the database segment were CRM and BI projects, PSUs, insurance firms, white goods manufacturers, FMCG and telecom companies, which embarked on huge BI and analytics projects. We expect the growth of RDBMS deployments to continue in India.

“The resurgence of the Indian economy post slowdown has resulted in a surge in information that has necessitated the need for a proper database management system. A greater need is being felt for better DB management tools for both structured and unstructured data. And since Indian firms are eyeing the overseas markets, this has also paved the way for the growth and deployment of DB tools. However, the adoption of these tools was confined only to large enterprises. But, in the last two years, mid-sized enterprises have shown increasing levels of adoption,” asserted Vohra.

Kathuria also pointed out, “India is a unique region with several small businesses rapidly growing and ramping up the technology they use to run their businesses. Microsoft, for example, was able to see a huge spurt in SMEs buying their SQL Server, and the number of unique customers buying their SQL Server has doubled in just six months and is seeing an increasing trend in adoption of its DBMS amongst smaller customers. The larger customers have already embarked on a solution deployment and are now at the cusp of reaping productivity from the solutions they have deployed.”

Operational benefits

The operational benefits of using RDBMS are manifold-the data is stored in rows, which makes it easier for large amounts of data to be stored in the database. This also makes the database readily scalable as the organization and the data within the organization grows. There are other benefits like robust security mechanisms, whereby if an organization intends to cordon off a part of the database
to a section of the employees, it could be done.

For example, by deploying Oracle Database 11g Release 2 as their data management foundation, organizations can utilize the full power of the database to reduce server costs by a factor of 5, reduce storage requirements by a factor of 10, improve mission critical system performance by a factor of 10, increase DBA and developer productivity by a factor of 2, eliminate idle redundancy in the datacenter, and simplify their overall IT software portfolio.

Oracle Database 11g also helps lower costs and delivers a higher quality of service by clustering and consolidating enterprise applications onto fast, reliable and scalable private clouds; accelerating performance by compressing data onto low-cost storage partitions; and securely protecting information and enabling compliance.

“RDBMS vendors of the current time provide high availability and DR solutions for the customer. This helps to maximize system availability and meet aggressive SLAs for system uptime, which is a critical operational parameter. Also, having an RDBMS can help consolidate several desktop databases and spreadsheets into a centralized database. This improves data sharing across the organization and makes it easier to manage. RDBMS also helps to reduce the redundancy of information as here the data is stored in discrete entities and is not spread all over as in case of flat files, which further helps to reduce the complexity in managing the data,” opined Tanweer Khan, Senior Project Leader, ADP India.

Pain points associated
with deployment

The adoption of RDBMS, especially the high-end data warehouse appliances, has a side effect of 'locking in' to a specific vendor. This is due to total dependency on the vendor for both hardware and software support. This, in turn, reduces the flexibility available to the customer to consider an alternate vendor later on. The RDBMS deployment strategy has to be reviewed in such event. Besides, migration of legacy application into RDBMS can also pose some challenge at times, which can be mitigated by technical support from the vendor.

There are other pain points associated with RDBMS like storing data in tables, which needs to communicate with each other. The storage and retrieval of data from an array of tables can be time consuming and might affect the performance of these tools. The data tables in RDBMS must be ordered, and it would be difficult without an 'index' to retrieve data. Also, the data is stored in a two-dimensional normalized format, which makes it difficult to understand all the relationships between tables and data. Another issue with RDBMS is that the data is stored in rows, which makes it difficult for use in conjunction with a data warehouse solution, where data is written only once, but needs to be accessed multiple times.

Moreover, there are advances in database management systems, like the evolution of column-based database systems, and object-oriented database management systems. Column-based database systems are faster than row-based database systems, and consume less space. The need for indexing, and storing the data in an ordered format can be avoided using object-oriented database management systems.

Evolving in the future

The RDBMS platform would continue to be the preferred DBMS platform in the coming years. Technologies like column-based databases, and object-oriented database management systems are relatively new, and would take some time to find acceptance in the India market.

Going by industry estimates, RDBMS will also move from being storage and retrieval mechanisms to decision making and elastic cloud-based offerings. Besides, the RDBMS vendors would continue their focus on providing more advanced high availability offerings. Enhancements in database manageability aspects will also continue to be focus areas for RDBMS vendors.

NIVEDAN PRAKASH
(nivedanp@cybermedia.co.in)