Gartner Study on DBMS Identifies Spending and Deployment Trends

In a recent Gartner study focusing on data management systems (DBMS) technology, investment and use, respondents provided information on their spending intentions and DBMS deployment characteristics. Respondents indicated historical and future spending for DBMS, as well as provided data on their plans to allocate their DBMS workload across operating system and open-source software (OSS) platforms. The results of this data indicate several key global trends, as well as highlight some critical differences among the regions.

Key Findings

  • Investment in DBMS software will increase across all regions in 2006.
  • North America will see the greatest increase in DBMS software spending.
  • Organizations will transition their DBMS workload across operating system platforms to achieve standardization and improve operational efficiency.
  • Not surprisingly, most organizations are moving some of their DBMS workload to Linux; however, z/OS on the mainframe is also a popular target.
  • DB2 will show strong growth during the next 12 months – this is greater than all other DBMSs.
  • Most organizations are beginning to research and adopt an OSS DBMS strategy.

Predictions

  • Through 2010, the DBMS software market will continue to see strong growth worldwide, driven primarily by new business intelligence (BI) and data warehousing implementations (0.8 probability).
  • By 2008, MySQL will be a serious choice for mission-critical applications (0.7 probability).

Recommendation

Despite the maturity of the DBMS market, ongoing focus and investment in DBMS software continues to be a competitive necessity and is fundamental to success in data-intensive initiatives, such as corporate performance management (CPM), BI and regulatory compliance. Organizations must recognize this trend and respond accordingly by aligning investment and deployment strategies for DBMS products with corporate goals.

ANALYSIS

In late 2005, more than 1,800 organizations worldwide responded to a Gartner questionnaire on data management. As part of the study, Gartner asked about spending plans, product selection and deployment of DBMS software.

2006 DBMS Spending Plans

When compared with 2005 spending on DBMS software, more than 70 percent of respondents indicated that they intended to increase spending on DBMS software in 2006. Nearly 50 percent indicated their spending would increase by more than 5 percent. These intentions reflect the foundational nature of the DBMS, because it supports all critical aspects of business operations and growth (see Figure 1).

Figure 1. Worldwide Spending Plans, 2006

Source: Gartner (March 2006)

Several critical differences in 2006 spending plans among the regions reflect the relative importance of data and data-related initiatives, as well as the general economic environment within the region. North America, the largest market for DBMS software, showed the greatest intent to increase DBMS spending in 2006, with 86 percent of survey respondents. Fifty-eight percent of respondents indicated that they intend to increase DBMS spending by more than 5 percent, and 30 percent indicated they intend to increase spending by more than 10 percent over 2005. This data reflects the importance being place on data-related initiatives, such as CPM and BI, as well as the general economic environment and a positive business outlook for the region.

This is an interesting contrast with Europe, which is the next-largest market for DBMS software. Sixty-nine percent of respondents intend to increase DBMS spending, but only 40 percent intend to do so by more than 5 percent. Europe contrasts with North America even further, where only 18 percent intend to increase spending by more than 10 percent compared with 2005. Europe and North America generally react similarly in their perception and usage of data for BI and CPM, so the difference in spending plans highlights the effect that the economic environment plays on organizations' spending plans.

Per Global Insight, Gartner's reference for economic data and forecasts, Europe is projected to see growth of only 1.9 percent in gross domestic product (GDP) in 2006. This contrasts sharply with a projected 3.5 percent GDP growth for North America. As a result, organizations in Europe show slightly more conservatism in their spending plans, reflecting a somewhat slower-growing economic environment.

Respondents in the Asia/Pacific and Latin America regions show similar spending intentions as Europe, however, for different reasons. In both regions, more than 60 percent of respondents indicated they intended to increase DBMS spending in 2006, and between 47 percent and 48 percent indicated that they intend to increase spending by more than 5 percent. However, both regions have projected GDP growing at approximately 4 percent, according to Global Insight. Hence, the lower spending plans for DBMS software, as compared to North America, don't reflect the economic environment (as is the case in Europe). Instead, the reduced spending plans reflect the relative importance of data-related initiatives in those regions.

One of the main drivers regarding the use of data to run the organization is the increasingly large amounts of data being generated from ERP and CRM applications. Because these types of applications are not widely used in the Asia/Pacific and Latin America regions, companies are not dealing with the volumes of data these applications generate. Hence, organizations in these regions have not seen a significant need for large-scale BI and data-warehousing projects to capture and analyze the data, and they don't have the impetus to invest as heavily in DBMS software.

Figure 2. DBMS Spending Plans by Region

Source: Gartner (March 2006)

DBMS Operating System Platform Plans

Many organizations are starting to view the efficient and strategic use of data as a critical competitive advantage, and are putting increasing focus and funding on BI and other data-intensive business initiatives. As a result, organizations are increasing purchases of DBMSs to support these data-intensive business initiatives, and, at the same time, many are taking this opportunity to standardize on a DBMS platform or platforms. As a result, organizations are moving some of their applications and DBMSs from older operating systems to Linux.

The survey results on where DBMSs are platformed are shown in Figure 3. Most respondents are using a Windows DBMS, which is clearly the result of large deployments of Microsoft SQL Server at the departmental level. This number should increase as SQL Server 2005 picks up market steam (released generally on 7 November 2005). Microsoft will continue to gain more than 50 percent of the new license revenue on the Windows platform. With 39 percent today and 20 percent in the next 12 months, Linux is gaining rapidly, and with 59 percent (compared with 65 percent in Unix), this demonstrates that Linux will overtake Unix in the next few years.

Figure 3. DBMS Platform

Source: Gartner (March 2006)

The 14 percent of new z/OS would equate to the number of new DB2 for z/OS installations being added annually. Many IBM z/OS installations are still using the pre-relational DBMS engines and are slowly converting these to relational technology on z/OS and other platforms. This trend will continue during the next five years and accelerate as more of the older applications on pre-relational technology reach the end of their useful lives.

The big news is that more than 50 percent of respondents will be making a platform decision on Linux during the next 12 months. Old and new Linux organizations are moving old applications or adding new applications to the Linux platform. As we have predicted, Linux will be used increasingly for simple and complex workloads with DBMS during the next five years. Linux is rapidly maturing as a mission-critical environment for applications. This can be seen in Figure 4, in which the largest growth is in Linux.

Figure 4. DBMS Platform Transition Plans

Source: Gartner (March 2006)

Although it's not surprising that more than 50 percent of respondents are making the transition from some of their DBMS workloads to a Linux platform, some of the other moves respondents plan to make are rather unexpected. Many respondents (20 percent) intend to transition some of the their DBMS workload from Windows to Unix, to standardize on a combination of Unix and Linux and improve operational efficiency. Another reason for moving to Unix is the perception that, with Windows, they are locked into a propriety operating system, and, with Unix, they are perceived to be more portable.

A surprising result was the nearly 30 percent of respondents who intend to transition some of their DBMS workload onto z/OS, a large percentage of whom are making a transition from Windows to z/OS. This reflects the robust reputation of the mainframe hardware, and the large installed base of mainframes. Many organizations view the mainframe as a "sunk cost" and are looking for areas in which to leverage past investments. DBMS is one area in which organizations can leverage the power and throughput of the platform through the consolidation of systems and head count. This is further illustrated by the low number of respondents (7 percent) who indicated that they intend to migrate some of their workload off of the z/OS platform.

We continue to see the mainframe chosen as older applications are rewritten or purchased. Finally, a percentage of those moving to Linux are moving to a zLinux LPAR on the z9 Series hardware for the same reasons of resource reduction. Many of these shops are moving Oracle from Unix to zLinux, often consolidating hundreds of Unix servers to a single mainframe.

DBMS Deployment Plans

Gartner used a series of questions to determine which DBMS engines were being used and analyze the impact of OSS DBMS. First, the distribution was as expected. In Figure 5, the widest deployed was Microsoft SQL Server, with 81 percent installed or planned. This is due to the widespread use of Microsoft SQL Server at the departmental level. Almost every organization uses SQL Server to some degree at this level. The 15 percent planned installations are due mainly to the new version SQL Server 2005.

Figure 5. DBMS Deployment

Source: Gartner (March 2006)

Oracle is the most widely installed production DBMS for enterprise applications, so it's not surprising that 74 percent of respondent's have Oracle installed now or planned for the next 12 months. Teradata also shows good growth, with 15 percent planned installations (it also shows consistent results across all regions, demonstrating a strong worldwide presence).

The biggest surprise is DB2 from IBM, with 19 percent planning to install DB2. This is nearly 50 percent higher than the number planning to install Oracle. DB2 has long been a strong DBMS offering, but has suffered from multiple versions, perceived lack of portability and a lack of third-party support. IBM has been marketing heavily during the past several years to change this perception and encourage software vendors to port to DB2. It would appear their efforts are working.

Even more interesting is that IBM DB2 is at 27 percent planned in the Asia/Pacific region, and Oracle is only at 10 percent. This is partially due to Oracle's broad penetration in this market, but also because of IBM's strong marketing efforts. The overall numbers so strongly in favor of DB2 indicate a pending increase in IBM's market share for DBMS during the next few years.

The most important trend indicated by the study is OSS DBMS. The MySQL OSS DBMS is installed in 49 percent of the responding organizations and is second only to DB2 on the planned installations, with 17 percent. Most of these installations are in the experimental phase, with some used in simple production environments. MySQL version 5.0 was released last fall and is the first release with full SQL compliance and the necessary functionality to support full applications. Therefore, its surprising strength so soon after release shows the strong worldwide desire to begin to use OSS DBMS systems. During the next several years, this DBMS will attract much attention and contend seriously with the other DBMS engines for application platform choice.

Figure 6 shows that, worldwide, 40 percent of the respondents are using or are planning to use an OSS DBMS. This percentage equates to 12 percent of the total respondents to the study. Another strong OSS indicator is that 42 percent of those respondents using or planning to use an OSS DBMS are researching the use for mission-critical applications of OSS DBMS (see Figure 7). We have been predicting that OSS DBMS will be ready in two to three years for mission-critical applications, and this is completely in line with these predictions. The implementation of OSS DBMSs for simple applications is accelerating, and the number of organizations researching, testing and deploying is increasing at a high rate.

Figure 6. OSS DBMSs Replaced by Proprietary DBMSs

Source: Gartner (March 2006)

Figure 7. OSS DBMS Usage

Source: Gartner (March 2006)

All of these OSS metrics demonstrate a clear movement toward OSS DBMS as a realistic platform. A growing number of organizations are starting to experiment by testing and moving them into production. By 2008, MySQL will be a serious choice for mission-critical applications (0.7 probability).

Acronym Key and Glossary Terms
BI – business intelligence
CPM – corporate performance management
DBMS – database management system
GDP – gross domestic product
OSS – open-source software

Gartner RAS Core Research Note G00138213, Colleen Graham, Donald Feinberg, 29 March 2006.

top

© 2006 Gartner, Inc, and/or its Affiliates. All rights reserved. Reproduction of this publication in any form without prior written permission is forbidden. The information contained herein has been obtained from sources believed to be reliable. Gartner disclaims all warranties as to the accuracy, completeness or adequacy of such information. Gartner shall have no liability for errors, omissions or inadequacies in the information contained herein or for interpretations thereof. The reader assumes sole responsibility for the selection of these materials to achieve its intended results. The opinions expressed herein are subject to change without notice.