By Nimish Shelat, Product Marketing Manager – HP Cloud & Automation
IT has come a long way in its ability to serve the needs of business users. Just consider the significant strides that have been made with the self-service delivery of servers, storage and network connectivity via internal cloud. But while these measures have helped improved responsiveness, service level owners and end users still often experience significant elapsed times between their service request and delivery. So what are the sources of delays now?
As many companies have discovered, provisioning “ping, power and pipe” via internal clouds has only shifted bottlenecks up the stack into the application layers. Web servers, application servers and databases each require a unique operations lifecycle that is not dominated by standard provisioning and configuration management, which is where most IT organizations focus data center automation efforts. Database and application admins often have their own slate of mundane, repetitive tasks that can have the effect of restricting the full agility of cloud-based IT.
That’s why well-designed and well-implemented data center automation must extend beyond the infrastructure layer to also include the application layer. The ability to provision and manipulate discrete database services is key to being able to support the rapid deployment and mass management of applications in the cloud.
What it really takes to manage application databases
Why can an IT organization do a great job offering server cycles or storage space in the cloud, but a seemingly lousy job efficiently managing databases and applications?
At the server layer, as well as storage and networking, significant admin time is required for provisioning. Even subsequent activities such as patching and configuration management are still often handled via provisioning—i.e., by reimaging the OS with a patched or reconfigured image.
But as you move higher in the stack, each discrete application layer requires a different use of an admins’ time. In fact, provisioning typically takes up just 15-20 percent of an app admins’ average workday. The remaining 80 percent of time is spent on other post-provisioning activities such as maintenance, incident response and service requests.
One reason is that the App Admin can’t simply provision a new image as one can with server administration. Instances of an application component frequently develop a unique fingerprint over the course of their use. Re-imaging an application server to apply a new patch from the Sys Admin wipes out much of this dynamic application fingerprint, creating a lot more work for the App Admins to restore it to the best extent possible.
Too many tools
Many tools exist for monitoring and alerting, or for carrying out ad-hoc and one-off tasks. Many do not provide the flexibility of using the command-line—which many DBAs still prefer—nor do they address actual automation. Propagating a new user and privileges to a couple dozen or even 100 databases isn’t possible with administrative GUI tools. Now add in a variety of mundane/repeatable tasks—installs, configs, refreshes and cloning, maintenance plans, patches, upgrades, migrations, managing load processes, DDL releases, and so on.
But don’t these tools have automation features? Sometimes, but they typically rely on triggering a script, which still needs to be written and tested by a time-pressed Senior DBA. Not only are other DBAs usually not familiar with it, but they must be documented centrally, and any updates to the scripts have to be manually logged across myriad database servers. Inevitably, errors creep in.
None of this positions IT teams to execute effectively on initiatives like Cloud-based Applications.
Automating more than infrastructure
HP addresses these common challenges in database administration by going beyond just monitoring and ad-hoc administration to actually improve the operating efficiency of your IT environment. HP’s broad automation platform offers out-of-the-box content to address the entire administrative lifecycle of databases (Figure 2).