As state and local governments’ hybrid, IT environments continue to grow and become more complex, is there a database skills gap that is growing alongside them? According to a survey of database professionals conducted by Unisphere Research in 2020, 84% of respondents claimed that they had two or more database platforms in use, while 31% said they had four or more. That’s a clear indicator of the growing complexity of hybrid IT with multiple database environments and the pressure facing database professionals who continue to be asked to do more with less.
Virtually gone are the days when a database professional could focus his or her attention on being “the expert” in one database system. A concerning skills gap arises when database professionals lack experience and specific training on the databases in their hybrid IT environment.
What are the challenges when a database professional needs to support a new database?
There are many reasons why supporting a new database is difficult. Although the general concepts are similar, the processes and procedures are different.
Processes that are common, ongoing and typically not time-sensitive include:
- Installing and configuring the database
- Backing up and recovering data
- Executing SQL and database-specific procedural SQL (e.g., T-SQL or PL/SQL)
- Using the GUI versus the command line for day-to-day tasks
- Setting security options and administration
- Establishing standards
Those with higher urgency can include:
- Troubleshooting problems
- Monitoring for availability (up/down time)
- Monitoring for performance bottlenecks at the database, server OS or virtualization level
- Investigating and diagnosing performance-related application anomalies or chronic issues
- Determining corrective action based on monitoring and diagnostics
- Convincing others in IT and application development teams what actions are required
Are there ways to overcome those gaps of skills and knowledge?
Since so many database professionals are now being asked to support databases outside their realm of expertise, there are ways to overcome their lack of practical experience and specific knowledge, including:
- Self-learning: Google searches, books, online courses, user forums and blogs
- Asking for help
- Mentorships
But these take valuable time that is often consumed by urgent tasks, especially if business-critical applications use the database in question. Finding someone in the organization with the skills and the willingness to act as the helper/mentor can be difficult or impossible. And none of those learning methods alone prepares the database professional for the pace they must maintain to keep databases running smoothly.
How can organizations help database professionals master the skills they need for these expanded responsibilities? One of the best ways is through translation of skills and knowledge from one database to another.
Make tasks relatable through translation
Many in IT feel that the best way to learn a new technology – in this case a database – is to make the new process relatable to what the individual has done before. So, if someone understands how to backup and recover data in one database platform, learning how to do it in a new platform won’t require complete reskilling, it will just require understanding the similarities and differences of doing it in the new database.
There are courses that take this approach, such as “Oracle or the SQL Server DBA” or “SQL Server for the Database Professional.” There are many options for Oracle and SQL Server. Less common, but needed now are courses that crosswalk PostgreSQL or MySQL with Oracle or SQL Server or “MongoDB for the Database Professional,” etc.
The translation approach works well for the administration, maintenance and even security aspects of the database professional’s job. Command syntax, configuration parameters, logging and transactions, separation of duties and database connectivity can be discovered and mastered pretty quickly.
What does not transfer quite as easily are the skills required when a database professional is under time pressure or in a crisis situation. That requires tools.
Foglight bridges the gaps, across a wide array of databases
As hybrid IT infrastructures grow in complexity at most organizations, with mixes of traditional relational databases, open source databases, NoSQL databases and cloud data warehouses, a tool that can help across that entire spectrum is needed. That tool is Foglight® by Quest®.
I listed some tasks earlier in the post and separated them into two categories. Since I feel there is quite a good chance that skills in the first grouping, “common, ongoing,” can be learned relatively quickly through the translation method, look back at the tasks in the “higher urgency” category – those tasks that a database professional must perform well at a moment’s notice. These tasks can be tackled using Foglight, no matter the skill level of the database professional on a particular platform:
Often, troubleshooting means “look at the logs” in response to a problem with the database itself – real or perceived. Maybe error messages are appearing when people run queries. Or maybe application screens are just stopping. In an unfamiliar database, finding and examining the log files might not be the right thing to do – partly because administration tools or vendor-supplied tools might not provide help there. Foglight monitors database and OS event logs for pre-filtered message text strings. If one of those messages (of high concern to you) appears, an alarm can fire. And, Foglight provides views of those messages within its web interface – the same interface where other diagnostic and historical information can be examined.
Availability is paramount, of course – if the database or a service isn’t up and running, nothing else really matters. If a standby database is out of service, or a database is unavailable for connection, a database professional must receive and instantly interpret this type of urgent information. Foglight provides up/down monitoring within the same interface as all other monitoring information, so immediate attention is possible – hopefully before the problem affects users.
Performance bottlenecks can be caused by a multitude of items internal and external to a database, and they occur frequently. Sometimes, bottlenecks are immediate problems that an alarm from a monitor can help detect. Other times, they are brought to attention after the fact, usually in the form of a complaint from a user along the lines of, “We had really bad performance on Monday afternoon. What’s wrong with the database?” Investigating problems days or even weeks later requires immediate access to historical information. Foglight has strong capabilities in collecting, storing and presenting this information so that someone new to the database doesn’t need to run commands at the command line or write scripts to extract monitoring elements. Investigations can take advantage of tools that provide important insights by correlating changes to the database to times of poor performance, or by comparing workloads and performance between timeframes.
Determining corrective action is sometimes not as urgent as the other tasks on this list, depending on the nature of the problem. If it’s an immediate problem affecting users of business applications, then immediate action is needed to minimize the impact on the business. Foglight quickly helps locate offensive SQL statements and users, allowing tuning efforts to focus on the most likely candidates for improvement. And the unique views built into Foglight’s web interface visually depict where bottlenecks are occurring, or have occurred, providing solid support for where to perform tuning.
Convincing others is an ongoing challenge, as database professionals must prove causes of performance issues – including whether the cause was on the database or elsewhere. In traditional IT silo “war room” scenarios after an outage or critical performance issue, proof always helps an argument, and Foglight provides excellent reports and visualizations that document the condition of the database at the time of a problem. If the database was the cause, Foglight’s forensic information can help determine the best course and appropriate people to involve. Then, Foglight becomes a collaboration and communication tool for sharing relevant information across hybrid IT teams.
This article originally appeared on Quest Software’s official blog, HERE.
