Wondering how many engines to allow? How to scale the Publisher cluster and identify bottlenecks? What configuration fine-tuning we have available under the hood?
Or really just wondering what the QlikView Publisher is at this point? Yeah, that's fair. Let's start with that:
The QlikView Publisher is the Reload & Distribution engine in a QlikView deployment. Using QlikView Publisher, you can automate the generation of data stores to be consumed, the reload of fresh data and the distribution of QlikView documents via email or to a specific QlikView Server or directory folder. In the QlikView Management Console, the Publisher is referred to as the QlikView Distribution Service, which is also the name of the Windows service managing the role. As a service, it can be clustered over multiple Windows server nodes.
And as the deployment grows and demand rises, you might start asking yourself a bunch of questions on how to address bottlenecks, prevent the same, or generally scale it better.
We here at Qlik are often approached by system administrators wanting to streamline their environment and address resource related issues that are currently causing their tasks to fail or not to trigger as expected. Fortunately, the QlikView Publisher is flexible and allows for a wide range of configuration options to help with that, though not all of them are immediately evident.
This new whitepaper (Scaling the QlikView Publisher) aims to demystify the scaling and fine-tuning of your Publisher cluster and provide you with basic best practices. It's been put together by Qlik Architects and Support Engineers and offers a good baseline to get started with, all of which can be further expanded on to fit your environment's specific needs.