If you’re like many DBAs, your life is probably getting more complicated by the week. Are you under pressure to manage larger and larger numbers of database instances? Or to adopt new database technologies like open source or NoSQL? With those changes, are you still required to ensure the best performance and stability for your entire environment?
A survey by Unisphere Research1 showed:
Many DBAs have had to simply dive in and face the challenges as they come. Each situation is unique, but I’ve identified the top five most common complications in this new era of having to manage cross-platform database environments.
Thanks to new development projects, mergers and acquisitions, and a range of other business scenarios, organizations are adopting more databases of various types. Not only are these databases being added, but DBAs are also being asked to manage them. The challenge here is that many DBAs consider themselves experts in one database platform and may try to apply that hard-earned expertise to newer platforms they are less comfortable with. Technologies, however, do not always translate. For example, SQL Server does not behave the same way as MongoDB in every situation, nor does Oracle behave the same way as PostgreSQL.
To adequately monitor a cross-platform environment, it’s important to first understand the basics needed to ensure database availability. Once general availability and disaster recovery are achieved, only then should you advance to more complex tasks, such as performance tuning, maintaining high availability and ensuring security. Database monitoring solutions can help with these challenges; however, monitoring can also be complicated because each database platform has its own methods, tools and APIs. The learning curve can be enormous.
Many database vendors offer several editions of their databases, such as a full-featured enterprise edition, a lower-cost standard edition and a stripped-down express edition. These same vendors often also offer database monitoring tools with some great features, but the tools frequently require licenses for the most expensive enterprise edition as well as a separate fee for the monitoring tool itself. With multiple platforms, this can get cost prohibitive pretty quickly.
With limited diagnostic capability, DBAs have fewer tools for resolving problems and less control over their databases — and a lot more frustration.
Most DBAs say the number of databases in their purview is increasing. More databases means less time to spend managing and troubleshooting issues on each one. Many large environments require that DBAs move from management of individual databases to inventory and capacity management of the entire environment all at once. It’s a balancing act that becomes increasingly more difficult as the environment grows in size — and it’s often fraught with more risk and potential downtime.
Besides being located on premises and in the cloud, databases can reside on physical or virtual servers. In global organizations, they can be anywhere on the planet, and DBAs must manage them remotely. Remote data centers can be necessary for distribution of data, more precise and up-to-the-minute business intelligence, and greater scalability. But DBAs must efficiently monitor and diagnose the performance of these far-flung database environments to keep all systems running smoothly 24/7. It’s kind of like octopus wrangling.
Clearly performance monitoring is key for cross-platform database management, but often the monitoring tools themselves can put a drag on performance. The main culprit is typically a misuse of a high-overhead API for pulling performance data out of a database server. Each platform has its own set of APIs, and many API’s have complex configurations. Choosing the wrong API or misconfiguring a specific option can quickly create troublesome overhead, and possibly even lead to downtime.
What to do?
It would be awesome if DBAs could simply apply their existing knowledge to every new platform that they must manage and to let that ripple out into the cloud and around the globe, keeping a constant, centralized view on things as their database environments continue to grow and change. But it’s just more complicated than that.
There is hope, however. Foglight for Databases can help DBAs meet all five of these key challenges. This one tool provides deep insight into the health and performance of multiple databases, regardless of platform, number or location.
For a more in-depth look at these challenges and how Foglight can help, download my tech brief, Managing cross-platform database environments, written with Pini Dibask, product manager for database management at Quest.
Or watch our Top 5 challenges of managing Cross Platform Database Environments on the same subject.
King, Eliot, “The Real World of the Database Administrator,” Unisphere Research.