the-real-world-of-the-database-administrator-white-paper-15623

Sponsored by
THE REAL
WORLD OF THE
DATABASE
ADMINISTRATOR
By Dr. Elliot King, Research Analyst
Produced by Unisphere Research,
a Division of Information Today, Inc.
March 2015
Produced by

2
THE REAL WORLD OF THE DATABASE ADMINISTRATOR was produced by Unisphere Research and sponsored by Dell Software. Unisphere Research is the market research unit of
Unisphere Media, a division of Information Today, Inc., publishers of Database Trends and Applications magazine and the 5 Minute Briefing newsletters. To review abstracts of our past
reports, visit www.dbta.com/About_Us#Unisphere. Unisphere Media, 121 Chanlon Road, New Providence, NJ 07974; 908-795-3702.
TABLE OF CONTENTS
Introduction and Key Findings��3
The Database Infrastructure��4
Data Under Management��10
DBAs’ Responsibilities��14
The Future��21
Conclusion��25
Appendix A��26

3
INTRODUCTION AND KEY FINDINGS
In the world of IT, perhaps nothing is more exciting than the
introduction of radically new technologies. And the information
management arena may be on the verge of one of those moments.
The past 10 years have been a golden age for generating and
capturing a virtual avalanche of new kinds and quantities of data.
With the growth of cloud computing and associated storage
technologies, social media, low-cost video, and other hot new
technologies, companies have access to an abundance of new
data types.
These developments have led to new ways of thinking about
data. Approaches like those represented by Hadoop and NoSQL
seemingly point the way to a post-structured-data world. And
concepts like data lakes paint a picture of assembling huge
pools of highly varied data that can be accessed and analyzed on
demand. The idea is to get to better data analysis, quicker and
less expensively using much larger datasets to advance everything
from fraud detection to predictive analysis.
The ability to capture, store, retrieve, analyze and save data
in new ways holds the potential to gain new insights and new
ways of doing business. If information is power, to paraphrase
an old adage, more information can lead to more power. Great
information can guide companies to greater business success.
But the attention to the new developments in information
management should not obscure the fact that structured data
in relational databases still provides the foundation for the
information infrastructure in many, if not most, companies, and
will do so in most organizations for the foreseeable future. The
ways in which companies can exploit structured data is far from
exhausted. And database administrators (DBAs) still represent the
front-line personnel for data management in most enterprises.
To understand the current role of the DBA and the way
that role is changing, Dell commissioned Unisphere Research,
a division of Information Today, Inc., to survey database
administrators and others charged with managing corporate data.
The 300 respondents came from a wide range of companies in
terms of size and industry. Nearly two-thirds of the respondents
came from organizations with more than 1,000 employees and
more than a dozen industries were represented. One-quarter of
respondents’ organizations are running more than 500 databases.
Details about the respondent pool can be found in the appendix.
Among the key findings are the following:
n While Hadoop and NoSQL are exciting new technologies,
their use currently is confined primarily to large companies.
The traditional database management system still
provides the foundation for the information management
infrastructure in most organizations. Oracle and Microsoft
SQL Server are the most common platforms to support
mission-critical data.
n Most enterprises believe that more familiar “new”
technologies such as virtualization and cloud computing will
have more impact on their organization over the next several
years than “newer” emerging technologies such as Hadoop.
In fact, Hadoop and NoSQL do not factor into many
companies’ plans over the next few years.
n Information infrastructures are not static. Most companies
run multiple databases and are open to adding new database
platforms if there is a need to do so. The most common
motivating factor for adding a new database management
system is the need to support new analytical applications.
n Most DBAs are responsible for multiple database instances
from multiple vendors.
n Structured data remains the bedrock of the information
infrastructure in most organizations.
n While maintenance and performance are the top
responsibilities for most DBAs, security is becoming an
increasingly important item on their agendas. However,
currently, DBAs spend less time on security issues than they
do on supporting database development.
n The key challenge for DBAs is learning new technologies.

4
THE DATABASE INFRASTRUCTURE
Although a lot of attention has rightly been lavished on
emerging new information management technologies, mainstay
database management systems still play the critical role in the
information infrastructure. As Figure 1 shows, Microsoft SQL
Server and Oracle are found in the overwhelming majority
of organizations, with MySQL, IBM DB2, and MongoDB
representing the next most popular database management systems.
The appearance of MongoDB on this list provides evidence
for a growing acceptance of NoSQL technology, particularly
in larger companies. Around 70% of all MongoDB users are
running more than 100 databases and 30% are running more
than 500 databases. And, nearly 60% of the MongoDB users are
in companies with more than 5,000 employees.
While there is relative diversity in the use of database
platforms—more than 20 different brands were mentioned by
respondents—Oracle, Microsoft SQL Server, IBM DB2 LUW, and
SAP Sybase ASE were the dominant RDBMSs running mission-
critical data, with Oracle and Microsoft SQL Server clearly the
leaders. Indeed, approximately 78% of the respondents indicated
that they were running mission-critical data on Oracle and 72%
said they were using Microsoft SQL Server as a platform for their
mission-critical data. Almost one-quarter of the respondents
said that they had turned to IBM DB2 LUW for mission-critical
activities and almost 12% have opted for SAP Sybase ASE.
In contrast, roughly 6% of the respondents are using
MongoDB, the most popular NoSQL database technology in
this pool, for mission-critical data. As a whole, the survey results
provide a good barometer for the current state of the adoption
of NoSQL technology. NoSQL is beginning to penetrate larger
companies but is not yet routinely a platform of mission-critical
data. Moreover, as Figure 2 shows, many companies have not yet
factored NoSQL into their plans in the next few years.
Along the same lines, although Hadoop has justifiably
generated attention as a next-generation information
management, its use is still largely confined to large companies.
In fact, 60% of respondents who are currently using Hadoop are
running more than 100 databases and 45% are running more
than 500 databases. Moreover, approximately two-thirds of the
respondents using Hadoop are employed by companies with
more than 1,000 people.
Furthermore, as Figure 3 shows, Hadoop does not yet factor
into most companies’ plans. At this time, only 15% of the
respondents are actually using Hadoop (and around another 5%
are in the process of deploying it), and nearly 60% currently have
no plans to use it in the future.
Clearly, traditional RDBMSs shoulder the lion’s share of
data management in most organizations. And since more than
85% of the respondents are running Microsoft SQL Server and
about 80% use Oracle, the evidence is clear that most companies
support two or more DBMS brands. As Figure 4 shows, by
far, the most important reason companies use more than one
database platform is to support different applications, with the
need to support multiple user groups the second most common
reason. Supporting increased data volume was only the fifth most
important reason for adding a new DBMS platform.
Under the right circumstances, organizations are open to
adding new DBMS platforms. The most common reason given
for adding a new brand is the emergence of a new analytical use
case. But other factors can also motivate a company to add a new
DBMS as well, including the need to improve performance and
flexibility as well as to better manage costs. (See Figure 5).
Several key insights are reflected in these results. First, most
companies run multiple databases and are open to adding new
database platforms if there is a perceived need to do so. The most
common motivational factor to add a new database management
system is to support new analytical applications. This scenario
bodes well for Hadoop if it can deliver on its promise to make
analysis more flexible and less costly. At this moment, however,
the use of Hadoop and NoSQL technology is largely confined to
bigger companies and many companies have no plans to explore
these newer technologies at all.
The Role of the DBA
As the information infrastructure changes due to both the
growth of data under management as well as the introduction
of new technologies, including cloud-based computing, it could
be expected that the role of the DBA, long on the front lines of
data management, would change as well. In this survey, 75% of
respondents’ companies had between 1 and 25 people with the
specific title of DBA, and 15% had more than 25 people with the
title of DBA (although in many companies, people who have other
titles will perform tasks associated with database administration).
As might be expected, most DBAs are responsible for multiple
database instances. As Figure 6 shows, almost half of the DBAs
manage more than 25 database instances each, and almost 10%
manage more than 100 database instances each.
Not surprisingly, the number of database instances for which
each DBA is responsible is growing. (See Figure 7.) The number
of databases for which each DBA is responsible is decreasing at
only 5% of the respondents’ companies.
Not only are DBAs responsible for managing multiple database
instances, they manage database management systems from
multiple vendors as well. As indicated in Figure 8, approximately
70% of the respondents said that the DBAs in their organizations
t
t

5
were responsible for managing databases from at least two vendors,
and 7% of respondents said the DBAs at their companies were
responsible for managing databases from five or more vendors.
The growth or potential growth of non-relational data
management technology such as NoSQL and Hadoop raises
an interesting question. Will DBAs be responsible for non-
relational information management technology? Or, will a
different job classification emerge to claim stewardship of those
technologies? The answer to that question is not yet clear but the
indication is that over the long term, DBAs will be responsible
for non-relational information management platforms as well.
Figure 9 shows that roughly two-thirds of the respondents that
had deployed either Hadoop or NoSQL said that DBAs were
responsible for managing those technologies.
Drilling down, in 62% of the organizations that have
already deployed a NoSQL platform—but not Hadoop—
DBAs are responsible for managing the NoSQL technology. In
organizations that have deployed Hadoop—but not NoSQL—
only 52% of the organizations have DBAs managing Hadoop.
And, in companies that have both Hadoop and NoSQL
installed, DBAs are responsible for managing the non-relational
technologies 72% of the time.
Clearly, the long-term trends that DBAs have faced are still in
place. DBAs are the front-line data management administrators.
Each DBA is responsible for multiple database instances from
multiple vendors and the number of databases each DBA is
expected to manage is increasing. But, there is a twist. Over the
next few years, as Hadoop and NoSQL become more common
in the enterprise, DBAs are also likely to be the ones responsible
for the administration of non-relational data infrastructure
technologies.
Microsoft SQL Server 87.37%
Oracle 79.86%
MySQL 41.3%
IBM DB2 LUW 28.33%
Mongo DB 14.33%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 1: What database brands do you have running in your organization?

6
Already using a NoSQL database 9.66%
Currently deploying 11.72%
Will deploy in 1 to 2 years 17.59%
Will deploy in 3+ years 3.45%
No plans 56.21%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 2: Does your company currently have plans to adopt NoSQL database
technology?
Already using Hadoop 15.15%
Currently deploying 5.3%
Will deploy in 3+ years 3.79%
No plans 57.58%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 3: Does your company use or have plans to use Hadoop?

7
Supporting multiple applications 80.29%
Supporting multiple user groups 45.88%
Supporting multiple workloads 31.18%
Managing database licensing and 29.03%
support costs
Supporting increasing data volumes 17.56%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 4: What is the most important reason for using multiple database
platforms in your organization?
Supporting new analytical use cases 46.42%
Improving database flexibility 35.47%
Improving database performance 35.09%
Supporting increasing data volumes 32.45%
Managing increasing database 32.45%
licensing and support costs
Supporting unstructured data growth 27.55%
Consolidating data infrastructure 25.28%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 5: What are the most important reasons for adopting a new database
management system in your organization?

8
One 1.14%
2 to 10 27%
11 to 25 21.29%
26 to 100 28.52%
101 to 500 8.75%
500 2.66%
Don’t know 10.65%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 6: Approximately how many database instances does each
DBA manage?
Figure 7: Is the number of databases for which each DBA is responsible
increasing, decreasing, or staying the same?
Increasing 72.31%
Decreasing 5%
Staying the same 22.69%

9
Figure 8: How many database platforms (i.e., platforms from different vendors)
is each DBA responsible for managing in your organization?
One 30.62%
Two 41.47%
Three to five 20.54%
More than five 7.36%
Figure 9: Are the DBAs that are responsible for managing relational database
management systems also responsible for managing non-relational
systems (such as NoSQL and Hadoop)?
Yes 68% No 32%
(Only respondents who had deployed Hadoop or NoSQL technology)

10
DATA UNDER MANAGEMENT
What seems to be explosive data growth has been a constant
feature of information technology and management for a long
time. The sheer volume of data under management seems
staggering and yet what was once considered a lot of data is now
seen as trivial. The respondents in this survey manage a high
volume of data, with close to 30% managing more than 500TB.
Over the past 5 years, however, the raw growth of data has
been only part of the story. As importantly, more data types are
being captured, stored, and made available for analysis. Moreover,
companies have access to more data that they themselves did not
create. Those two drivers—new data types and new data sources—
have led to the interest in what is called big data.
For all the interest in how to capture and manage unstructured
and semi-structured data, structured data remains the bedrock
of the information infrastructure in most companies. As Figure
10 shows, two-thirds of the respondents indicated that structured
data represented at least 75% of the data under management.
Nearly one-third said that their organizations do not actively
manage unstructured data at all.
Interestingly, the reported rate of growth for the total amount
of data under management is substantially less than is commonly
reported. Slightly less than half of the respondents reported that
the total amount of data under management—both structured
and unstructured—has been growing at a rate of 25% or less
annually. Looked at another way, however, around two-thirds of
the respondents said their overall data is growing between 10%
and 50% annually. (See Figure 11.)
The growth rate of structured data generally parallels the
growth rate of data overall. As Figure 12 shows, about two-thirds
of the respondents indicate that their structured data is growing
between 10% and 50% annually.
Since some organizations are not yet managing much
unstructured data, the rate of growth for unstructured data
overall as reported in this survey is not as great as might be
expected. Figure 13 shows that less than 12% of the respondents
believe that the growth rate of unstructured data under
management in their organization is more than 50% annually.
But this chart also reveals another important point. About one-
third of the respondents simply do not know how fast the amount
of unstructured data under management is growing. This can be
taken as evidence that in many organizations, unstructured data
may still be outside normal data management processes.
Figures 14 and 15 show the sources for data growth. Figure
14 asked about the sources of structured data growth and Figure
15 posed the same question for unstructured growth. Both
returned interesting results. The primary driver for the growth
in structured data is transactional data, including e-commerce
data, followed by data from ERP systems and then financial data.
Structured data from the internet outside of e-commerce, such as
clickstream data, represented a significantly less important source
of structured data growth.
As for unstructured data, despite all the attention that has
been lavished on social media as a potential new data source
for analytics, the most important driver for the growth of
unstructured data is internally generated documents, followed by
email. Clearly, when it comes to unstructured data, the challenges
of managing internally generated data have a top priority.

11
Less than 10% 3.49%
At least 10%, but less than 25% 6.98%
100% (we don’t actively manage 28.29%
unstructured data)
Don’t know 9.3%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 10: What percentage of this data is structured (i.e., stored in traditional
relational database management systems)?
Less than 10% 7.39%
100% or more (i.e., the amount of data 2.33%
under management is doubling annually)
Don’t know 12.06%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 11: How fast is the total amount of data under management at your
organization growing annually (both structured and unstructured)?

12
Less than 10% 8.66%
Don’t know 11.81%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 12: How fast is the amount of structured data growing annually?
Less than 10% 17.39%
Don’t know 32.81%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 13: How fast is the amount of unstructured data growing annually?

13
Growth in transactional data 83%
(including e-commerce)
Growth in management data 51.78%
(such as ERP systems)
Growth in financial data 40.32%
Growth in structured data from internet 24.9%
sources (such as clicks of visitors to
your website)
Other 3%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 14: What are the most important sources of structured data growth
at your organization?
Growth in internally generated 50.41%
documents
Growth in email 35.54%
Other internet sources of data 30.58%
such as the web
Machine-generated data 31.4%
New data types such as video 29.34%
and audio
Social media 25.62%
Other 4%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 15: What are the most important sources of unstructured data growth
at your organization?

14
DBAS’ RESPONSIBILITIES
Given the importance of data and databases to the overall
functioning and success of every organization, effective database
administration is critical. DBAs perform many tasks, and the
demands on their time have changed as database technology has
evolved and new capabilities have been introduced. Some tasks
that were once performed manually are now automated. And
many database administration tools have been introduced that
help DBAs perform their essential tasks.
Figure 16 shows the tasks DBAs perform that respondents
consider the most important. The results are not surprising.
The DBAs’ top responsibilities are maintenance, performance,
and configuration. In other words, DBAs must make sure that
the databases are up and running, performing well, and are
configured to function appropriately. Beyond that, DBAs must
be concerned with a wide range of tasks ranging from capacity
planning to security.
The way DBAs spend their work days largely aligns with
their perception of their most important responsibilities. (See
Figure 17.) Interestingly, however, DBAs may spend less time
on security issues than they do on supporting development even
though security is generally a higher priority. They also seem to
spend less time on data integration and capacity planning than
might be expected.
Given that DBAs view systems uptime and performance as
their two most important responsibilities, it is appropriate that
those are the two metrics by which they measure themselves.
Figure 18 shows the top five criteria by which DBAs evaluate their
own performance.
Although the way DBAs judge their performance is consistent
with what they perceive as their top priorities, they believe that
their managers evaluate them differently. As Figure 19 reveals,
survey respondents felt that the speed of resolving problems
was a more important issue for managers in evaluating DBA
performance than system performance, and system performance
was slightly more valued than system uptime.
Of course, the way DBAs spend their time changes as new
responsibilities are added, as new technology is incorporated
into the infrastructure both at the data management and data
administration levels, and as staffing changes and organizations
generally evolve. Figure 20 presents some of the granular tasks
DBAs perform and assesses which of those tasks now require
more time and which require less.
For most DBAs, the amount of time spent on specific tasks
is not changing much. But there are some notable exceptions.
More than 60% of the respondents indicated that they are
spending more time on performance tuning and more than 53%
said cloning or provisioning databases for test and development
required more attention. Very few tasks seem to have required
less attention. Indeed, the only tasks tabbed by more than 20%
as requiring more attention were verifying that all scheduled jobs
had run (22.08%), and verifying all instances are up (20.6%).
One approach to making a DBAs’ job more efficient is to
automate routine tasks. There are several different approaches
to automating tasks, including using third-party tools or writing
custom scripts. Figure 21 provides insight into which DBA tasks
are automated.
Clearly, at most organizations, many fundamental
administrative tasks are automated. More than 80% of the
respondents say that verifying that all the database instances are
up has been automated. And more than 70% report that verifying
all jobs have run, backups have been successful, and there is
adequate disk space, is currently automated as well. The tasks
that are most commonly performed manually are performance
tuning (78.85%) and provisioning or cloning databases (67.98%).
That finding is consistent with the earlier finding that those tasks
require an increasing amount of DBAs’ attention.
Finally, as more tasks are automated DBAs are sometimes
called on to do database development tasks as well. In Figure 22,
respondents indicated which development tasks they thought
required the most time, regardless of whether development was
done by a DBA or a database developer.
Even as new non-SQL-based technologies begin to gain
traction, SQL is still very much the primary focus for most DBAs
and database developers.

15
Maintenance (backup, alerts, 70.16%
integrity checks, defragmentation, etc.)
Performance (system and data availability, 68.55%
diagnostics, optimizing and tuning, etc.)
Configuration (adding features, patching, 56.45%
upgrading, etc.)
Change management 44.35%
(deploying schema changes, etc.)
Security (auditing, configuring, 41.13%
documenting, etc.)
Data integration (ETL, moving data, 33.06%
merging data, etc.)
Capacity planning (forecasting future 33.06%
workloads and system capacity needs,
identifying new technologies, etc.)
Supporting development efforts 33.06%
(creating test and development databases
and synchronizing them with production, etc.)
0 20 40 60 80 100
0 20 40 60 80 100
Figure 16: What do you consider the most important responsibilities
for DBAs?

16
Maintenance (backup, alerts, integrity 63.11%
checks, defragmentation, etc.)
Performance (system and data availability, 59.43%
diagnostics, optimizing and tuning, etc.)
Configuration (i.e., adding features, 46.31%
patching, upgrading, etc.)
Change Management (deploying schema 44.67%
changes, etc)
Supporting development efforts 31.56%
(creating test and development databases
and synchronizing them with production)
Database development (data modeling, 27.46%
programming, deploying code to production, etc.)
Security (auditing, configuring, 22.95%
documenting)
Data integration (data management, 22.95%
ETL, moving data, merging data)
0 20 40 60 80 100
0 20 40 60 80 100
Figure 17: To which of these responsibilities do you think DBAs typically
devote the most time?

17
System performance 72.95%
System uptime 67.62%
Project delivery time 32.38%
Efficiency in resolving pre-production 23.77%
issues such as identifying and fixing bugs
Efficiency in resolving post-production 20.49%
issues such as configuration problems
0 20 40 60 80 100
0 20 40 60 80 100
Figure 18: What are the most important metrics by which you measure
your performance?
Speed of issue resolution 59.84%
System performance 54.1%
System uptime 52.05%
Adherence to service-level agreements 34.43%
Capacity management 9.84%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 19: What are the most important metrics by which your manager
measures your performance?

18
Figure 20: Among these responsibilities commonly assigned to DBAs, is
the amount of time you devote to each responsibility increasing,
decreasing, or staying the same?

Increasing Decreasing Staying the same
Verify all instances are up 24.89% 20.6% 54.51%
Inspect error logs for unusual events 29% 17.32% 53.68%
Verify that all scheduled jobs have run successfully 16.88% 22.08% 61.04%
Verify success of database backups 23.38% 19.48% 57.14%
Monitor disk space 30.3% 19.48% 50.22%
Review database size and growth settings 37.83% 12.61% 49.57%
Cloning or provision databases for test and 53.68% 10.39% 35.93%
development
Verify that replicated databases are synchronized 26.09% 13.91% 60%
Performance tuning for database workload 61.21% 4.31% 34.48%
(rewriting SQL statements, adding indexes, etc.)

19
Figure 21: Of the tasks that many DBAs perform, which are automated,
performed manually or not generally performed?

Automated Manual Not generally
performed
Verify all instances are up 81.14% 15.35% 3.51%
Inspect error logs for unusual events 46.49% 45.61% 7.89%
Verify that all scheduled jobs have run successfully 70.93% 23.35% 15.73%
Verify success of database backups 71.18% 25.33% 3.49%
Monitor disk space 72.49% 24.45% 3.06%
Review database size and growth settings 40.79% 50.88% 8.33%
Cloning or provision databases for test and 23.68% 67.98% 8.33%
development
Verify that replicated databases are synchronized 42.92% 36.28% 20.8%
Performance tuning for database workload 15.42% 78.85% 5.73%
(rewriting SQL statements, adding indexes, etc.)

20
Tuning SQL 73.48%
Writing and debugging code 39.57%
Editing and making code changes 32.61%
Querying 27.83%
Load testing 24.35%
Unit testing 11.74%
Re-factoring 10.87%
Other 3%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 22: Regardless of who is specifically responsible for database
development, which of the following development tasks
do you feel occupies the most time?

21
THE FUTURE
The relational database has been at the foundation of large-scale
information management for several decades and should continue
to be so for the foreseeable future. That said, interesting new
information management, storage, and analytical technologies are
making their way into many enterprises and DBAs will play a role
in their implementation and use. At the same time, as companies
use more data types from more sources, both complexity and risk
are escalating. Both of those developments will present challenges
to DBAs. Figure 23 reveals what respondents believe will be the
most significant challenges facing DBAs over the next 3 years in
terms of the developments in data itself.
Managing data growth and improving security are on most
people’s radars as areas that need to be addressed in the future.
Beyond that, there is widespread disagreement about where DBAs
need to focus over the next several years. This variety of opinions
can be seen as evidence that we are heading into a period of change
and the demands those changes will place on DBAs is not yet clear.
The same wide array of opinions exists when respondents
are asked to anticipate the key changes in the data management
infrastructure over the next several years. As Figure 24 suggests,
no single technology or change has truly come into sharp focus.
No single technology or changes was tabbed by at least 50% of
the respondents.
Interestingly, in terms of new technologies, DBAs do not
believe that Hadoop and NoSQL are going to have the biggest
impact on their organizations over the next several years. They
are more focused on the continued growth of cloud computing
and virtualization. (See Figure 25.)
Obviously, changes in the data and data management
infrastructure will put pressure on DBAs and their organizational
teams. As Figure 26 shows, DBAs realize that they will have to
learn new technologies and they may have to do so in an era of
smaller budgets.
Change, of course, represents both opportunities and threats.
DBAs will be critical for enterprises to be able to maximize the
benefits of the opportunities and minimize the threats.
Over the next year, however, the most pressing items on the
agenda of DBAs will be more evolutionary than revolutionary.
When asked to identify their most important project for next
year, the most common answer was upgrading their database
systems or enterprise applications, followed by consolidating
their information infrastructure. That consolidation could
come as the result of an upgrade or move to the cloud, or it
could be independent of other projects. Cloud technology is
on the radars of many DBAs. Many would like to automate
more tasks associated with database administration. No single
project or project area emerged as a consensus issue from this
survey. However, one of the most challenging aspects of database
administration is that each setting is unique.

22
The overall growth of data 66.23%
(structured and unstructured)
Improving data security 55.26%
Implementing databases running 37.28%
on cloud
Managing multiple database platforms 35.96%
The proliferation of new sources for data 28.51%
The need to manage new data types 25.44%
Improving data quality 24.12%
The overall growth of unstructured data 23.68%
The overall growth of structured data 20.18%
If other, please specify 3%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 23: In terms of data itself, over the next 3 years, what do you
anticipate will be the top challenges facing DBAs?

23
Incorporating cloud technologies 43.67%
Improving overall system performance 39.3%
Automating more tasks associated 38.86%
with database management
Increased complexity (heterogeneity) 38.43%
Incorporating non-relational data 37.12%
management technologies
Increasing overall system capacity 34.06%
Making data available to more 25.76%
stakeholders
0 20 40 60 80 100
0 20 40 60 80 100
Figure 24: In terms of the data management infrastructure, what do you think
will be the top challenges facing DBAs over the next few years?
Cloud 64.32%
Virtualization 48.02%
Big data 47.14%
High availability 33.48%
NoSQL 33.04%
Hadoop 29.96%
Consolidation 23.35%
IaaS (infrastructure-as-a-service) 21.15%
Agile development 19.38%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 25: What technology trends do you think will have the most impact
on database administration over the next 3 years?

24
The need to need to learn new 53.51%
technologies
Shrinking IT budgets 49.12%
The need to manage more databases 46.05%
per DBA
The lack of trained personnel 37.72%
Shrinking IT staffs 32.46%
Not having the proper database 17.54%
management tools
The need to train more users 8.77%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 26: In terms of database administration itself, over the next 3 years,
what do you think will be the top challenges facing DBAs?

25
CONCLUSION
While exciting new information management technologies
such as Hadoop and NoSQL are beginning to make inroads in
large companies, by and large, traditional database management
systems still represent the critical informational foundation in
most companies. DBAs are the front-line personnel maintaining
that foundation, responsible for ensuring that performance meets
the needs of the organization. With that in mind, several long-
term trends are still in place. DBAs are being asked to manage
more database instances and DBMSs from different platforms.
The amount of data under management continues to grow at
a crisp pace. And while some routine administrative tasks have
been automated, performance tuning, and provisioning are still
often done manually.
Perhaps as importantly, as the use of new technologies
grows, DBAs will be the ones responsible for integrating those
opportunities into their organizations. To that end, they have to
continue to learn new technologies and continue to find ways to
do more with less.

26
APPENDIX A
The following figures show the characteristics of the pool of respondents.
Database administrator (DBA) 48.03%
IT director or manager 19.65%
Database developer 4.37%
Analyst 3.49%
IT staff 2.62%
Other CEO 2.18%
Other C-level executive 1.75%
Project manager 1.75%
Chief information officer (CIO) 1.31%
DevOps engineer 1.31%
LOB director or manager 1.31%
Chief data officer (CDO) 0.44%
Other 11.79%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 27: What is your primary job title?

27
IT services/consulting/system integration 18.86%
Financial services 12.28%
Government (all levels) 10.96%
Utility/telecommunications/transportation 8.33%
Software/application development 7.89%
Education (all levels) 7.89%
Manufacturing 7.89%
Retail/distribution 6.14%
Healthcare/medical 5.7%
Insurance 4.39%
Business or consumer services 2.19%
High-tech manufacturing 1.75%
Other 5.7%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 28: What is your organization’s primary industry classification?
1 to 100 employees 11.35%
101 to 500 employees 13.97%
501 to 1,000 employees 10.48%
1,001 to 5,000 employees 24.45%
5,00 to 10,000 employees 13.1%
10,000 employees 24.02%
Don’t know/unsure 2.62%
0 20 40 60 80 100
0 20 40 60 80 100
Figure 29: Approximately how many employees (including all locations,
branches, and subsidiaries) are in your organization?

the-real-world-of-the-database-administrator-white-paper-15623

More Related Content

What's hot (20)

Viewers also liked (12)

Similar to the-real-world-of-the-database-administrator-white-paper-15623 (20)

the-real-world-of-the-database-administrator-white-paper-15623