4 Tips for DBAs Managing Multiple Databases

Bookmark and Share

Data is the lifeblood of businesses today, and there are more ways than ever to collect, store and extract value from it. But while new approaches to data management offer advantages there are also challenges for DBAs who are being tasked with overseeing these new and more diverse environments.

The DBA’s role is becoming increasingly complex and now often involves managing data both on premise and in the cloud, and in a variety of database management systems including not only traditional relational systems such as Oracle, SQL Server, MySQL and others, but also new open source and NoSQL technologies, according to Gerardo Dada, vice president of product marketing and strategy at SolarWinds.

In a recent interview, Dada offered tips for how DBAs can keep up with the pace of data in business, particularly as they are faced with the need to adopt new data management approaches.

According to Dada, today it is more important than ever for companies to:

  1. Have a common set of goals, metrics and SLAs across all databases, ideally based on application response times, not only uptime
  2. Document a consistent set of processes for ensuring integrity and security: backup and restore processes, encryption at rest and on transit, detection of anomalies and potential security events in logs, to name a few.
  3. Establish a strategy, roadmap and guidelines for moving to the cloud (or not) and for reducing workload costs by moving databases to lower-license-cost versions or open source alternatives
  4. Make sure team members can escape firefighting mode and spend enough time proactively optimizing the databases, which can result in significant cost savings

What Has Changed in IT Environments

“In today’s world where there are so many mergers and acquisitions and companies are relying more and more on software, DBAs are being faced with having to support more than one database engine in many cases. That is why this whole conversation is important,” said Dada. “We also see from research that DBAs and application teams are now having to manage more databases than before. That is why having more automation and having more thoughtful processes ahead of time is going to be important for them in order to be successful.”

In the past, making sure the database worked was the primary job of the DBA, but as DBAs have had to support more platforms than ever before, and organizations rely on software more heavily than before for everyday tasks - ranging from salespeople using CRM applications to shoppers use ecommerce sites - that is no longer acceptable. “They need to become more proactive to prevent problems,” Dada noted. Companies such as Need for Faster Application Response Times

Google and Facebook have set the bar on what is an acceptable response time for an application. Today, if an application is not responding to a user in 3 seconds, it is problem. “Today, an application that is really slow is effectively not up, and not working, because it is impacting the business,” said Dada. It is not enough today to have the database be up, it needs to be working fast. “The database is the heart of any application, and that means that if DBAs are going to have to manage SQL Server, Oracle and MySQL, they need to agree with management on what is going to be the focus of the effort and what is the acceptable response time that they are going to aim for, and how are they going to approach it so that there is less firefighting and more upfront investment in thinking about what needs to be done proactively.”

Avoiding Expensive Hardware Upgrades

There is another benefit as well to being more proactive, and to shifting the role of the DBA from managing for uptime to managing for performance, Dada pointed out. Companies today are making significant investments in hardware, in terms of server hardware, virtualization and storage. Many companies are planning to move off of spindle disks to flash storage for example and performance is a key driver of that. This can mean an investment of a couple of hundred thousand dollars or a move to the cloud. In both of those cases, performance is going to be critical and the database is typically the central element that impacts performance.

The DBA should not only be a partner in those conversations as things are being moved around and investments in hardware considered, but can also be a contributor to reducing the requirements for those hardware investments in the first place by making the databases run faster so in many cases the need for flash storage, for example, might not be as urgent.

And the converse can actually be worse, said Dada. If an application team makes the case for the business team to invest in faster storage and when it is done, they realize that there has been no benefit because the bottlenecks were not in the storage system in terms of writing and reading from the disk but in other parts of the application, they then have to tell their management that they just spent hundreds of thousands dollars and didn’t solve the problem. “That is an even worse situation.”

New Database Technologies Add Complexity

Increasingly, as more organizations are considering new database alternatives such as newer technologies such as open source databases such as MariaDB, MySQL, PostgreSQL, Percona, and NoSQL databases such as MongoDB and Cassandra, DBAs need to become more proactive in driving those decisions as well. The point, said Dada, is that the need to adopt newer technologies such as NoSQL when appropriate, and cloud, which is becoming a mandate in many IT departments or to adopt open source databases for cost savings affect DBAs. All three impact what the DBAs are working on. The DBA can get ahead of this pressure to move to open source, NoSQL, or cloud by determining in advance the pros and cons of each technology, which workloads will work better in the cloud versus on premise, and the elements of an application that work better in a NoSQL database versus those that need to remain in a SQL-based database.

And when adopting a cloud based approach, security and governance policies including encryption at rest and during transit and who has access to that data needs to be evaluated ahead of time. “We only need to look at the news to see the damage that has been done to companies by both internal and external threats.” Detecting and stopping data from being extracted from databases in real time must be considered in advance. Again, said Dada, DBAs must move from firefighting and take a step back to think about this before they move their first workload to the cloud or to a different database system.

New Database Technologies Coupled with a Skills Shortage

The maturing of cloud and NoSQL and big data approaches means that no CIO wants to be the one in their group of colleagues that has not looked at those technologies, and there is pressure on organizations to evaluate them and determine where they are useful.

“We have seen some IT departments that have gotten a directive of cloud-first,” said Dada.  The maturity of the cloud and NoSQL has gotten to the point where they are viable technologies and the same is true of open source databases.

The second trend putting pressure on DBAs is that there is a skills shortage. Even if a company wants to hire more DBAs for these newer technologies, it can take months to recruit the right person and get them in place, leaving the existing DBA team to manage with the resources they have in the interim, said Dada.

These two converging trends are placing more pressure than ever on DBAs, said Dada, and making it more of an imperative than before that DBAs become more proactive in tuning their database environments and proactively evaluating new technologies.

Related Articles

The managers and professionals who dedicate their careers to fundamental database administration are a shrinking pool. Many are retiring, while others have career aspirations outside of the IT department. The data profession itself is splintering into an array of new specialties and tasks—away from database administration and programming and toward higher-level data science and business consulting tasks.

Posted October 08, 2014