By Lingxiang Cheng, Senior Consultant
QlikView is a great business discovery tool that can easily generate excitement among business users. However, to build out a QlikView system that can support tens, hundreds, or even thousands of users is no easy task. As a QlikView consultant I have seen these scenarios many times:
- A business analyst is eagerly showing an IT manager the beautiful dashboards he/she builds in QlikView, and can’t wait to show the dashboards to hundreds of internal and external users;
- A business executive wants all business users to self-service their information needs through a QlikView platform.
Often times the IT managers seem to be less excited than their business counterparts. To be fair, in order to build a solid system, IT managers have a lot of questions that need to be answered. Let’s take a look at the most commonly seen ones (I have to warn you it’s a bit technical).
VMware for the servers?
Nowadays most IT infrastructure professionals are in love with Virtual Machines (VMs). With its lower cost and flexibility in terms of horse power, VM seems to be the way to go for all enterprise applications. Well this is almost true for QlikView. All components of QlikView services can work fairly well on VM except for QlikView Server. This is because QlikView Server does the heavy lifting in-memory data processing, and the extra VM layer is going to slow it down, a lot. Most seasoned QlikView consultants would recommend at least using a physical machine for the production QlikView server.
Separate Publisher from Server?
The two most important components in a QlikView system are the QlikView Server and Publisher (aka Distribution Service, I am a QlikView veteran so I will stick to the old name). Most people install all components on the same machine but as your data volume and user base grow, people frequently see these two components getting in the way of each other. At this point, it might be a good idea to put Publisher on a second machine (by the way, that second machine could be a VM). The good news is you can still manage both the QlikView Server and Publisher from a single QlikView Management Console (QMC).
QlikView Web Server or IIS?
QlikView comes with its own web server, which is a light version of Microsoft Internet Information Service (IIS). In a perfect world, where all users are Active Directory users, and the number of users is small, QlikView’s web server works pretty well. For a large user base deployment you might want to choose to use IIS because it’s more scalable. IIS also gives you more security options.
Security is always a big issue in an enterprise. Typically security includes authentication, which is about making sure a user is who he/she claims to be, and authorization, which is about letting the user see only what he/she is allowed to see. In the real world, we often see QlikView Publisher and Section Access can handle authorization pretty well, while the authentication is a bit more complex. This is because QlikView does not do authentication itself, it always relies on third-party authentication like NTLM, SiteMinder, WebSeal, etc. This process is also known as Single-Sign-On (SSO). This normally plays well in an enterprise because usually there is already a standard authentication product in place. The good news is, if the existing standard authentication product happens to be NTLM/Active Directory, QlikView can work with it out-of-box with no additional configuration. Otherwise, the third-party authentication product needs to inject the user id into http header and QlikView needs to be configured to read user id from http header. If the QlikView Servers are also accessed by outside people, it might be a good idea to setup a reverse proxy.
I hope by now you get a sense of the extra “behind the scenes” efforts that are necessary to make your QlikView system successful. In my follow up blog, I am going to talk more on topics such as high availability, storage, and running load tests.
About the Author
Lingxiang Cheng is a Senior Consultant for Bardess Group Ltd. He has been working business intelligence and data warehousing for over fifteen years. Lingxiang contributes his expertise to Bardess for solution architecture, development, and training. He has been working as business intelligence consultant and architect for many Fortune 500 companies in retail, pharmaceutical, financial services, insurance, and life sciences.