Getting from raw data to usable datasets is a non-trivial exercise, even with a powerful tool like Informatica at your disposal. Some estimates place the extract, transform, load (ETL) stage at as much as 80 percent of the entire process. As tempting as it is to call the job done once Informatica is working and yielding usable data, that would be short-sighted. A process that makes such heavy claims on your department’s time and resources merits all the attention and refinement you can manage.
If you’re a serious gamer, you’ll know that even modest improvements in your network throughput, or coaxing a few extra frames per second from your CPU and GPU, can have a dramatic impact on your gameplay. Aside from the in-game benefits, and bragging rights among your friends, there’s a real intellectual satisfaction to be had from the problem-solving process itself.
Diligent tuning of your Informatica implementation is just as rewarding as tweaking your gaming rig, and the gains in productivity are much farther-reaching.
Carrying out regular health checks is the starting point. Health checks address fundamental structural issues in your Informatica usage. They assess architecture and performance of your systems as well as the maintenance tools that have been baked in. Ideally you’ll be able to diagnose potential problems before they impact performance too greatly.
Start in the database, where your data actually resides. This includes your source and target databases, as well as any transient special databases where your data might be held during staging or transformation. A well-constructed health check looks at them all, monitoring details like the time required for popular queries, indexing and memory management issues, and potential conflicts between scheduled processes.
How often you run these checks is discretionary, depending on their importance. Some you might opt to run every day, while others might only require weekly, monthly or even quarterly execution. When your health check identifies a problem, you can bring your skills to bear on isolating the cause and creating a fix for it.
Once you’ve addressed or ruled out structural issues across your databases, conventional tuning approaches – testing the connection speeds between your servers, for example, and then tweaking your I/O settings accordingly – can also yield appreciable performance improvements.
Tweaking Informatica and its underlying databases and processes is a project with a large number of moving parts and an almost infinite range of potential interactions between them. The more experience you can bring to the table, the better you’ll be able to diagnose and manage performance.
This is where you’ll find value in the scale of real-world experience available through Datavail’s team. Our newest white paper, “The Essential Guide to Informatica Health Checks & Performance Tuning,” distills that collective experience into a few short pages packed with hard-won expertise.
With this direction at your fingertips, you’ll be able to take a structured, methodical approach to maximize the value of Informatica.
This blog post discusses the release of Oracle Analytics Cloud, a service that includes Essbase Cloud, Business Intelligence Cloud and Data Visualization Cloud.