The Google Cloud Platform (GCP) contains many components and capabilities that are directly relevant to a deployment of Qlik software. This non-exhaustive list catalogs some of the commonly used services. More detail can be found at https://cloud.google.com/products/
These may be accessed using the corresponding ODBC driver that may be downloaded from Google, or often using one of the drivers in the Qlik ODBC Connector package that is embedded in Qlik Sense.
Persistent Disk – durable and high performance block storage (SSD & HDD) that may be attached to instances of Google Compute Engine (Virtual Machines) or Google Kubernetes Engine (Containers) https://cloud.google.com/persistent-disk/
Cloud Storage (GCS) – Object storage with global edge-caching. May be used for CLOB artifacts such as extractions from BigQuery but not suitable as a fileserver. https://cloud.google.com/storage/
Compute Engine (GCE) – Virtual Machines running in Google data centers. This is the simplest step towards Cloud whereby you no longer manage Hardware, but you do still have to maintain (patch) the Operating System and manage all applications on the host. QlikView and Qlik Sense Enterprise for Windows require Compute Engines. https://cloud.google.com/compute/
Kubernetes Engine (GKE) – Containerized Application Management. Containers are a step beyond Compute Engines, in which you no longer maintain the [typically Linux] Operating System, but only configure how many Containers your Application runs in. Qlik Sense Enterprise for Elastic runs in Kubernetes-managed Containers. https://cloud.google.com/kubernetes-engine
Cloud VPN – extend your on-premises network to Google’s network through an IPsec VPN tunnel. Useful for making your on-premise Active Directory available in your VPC or providing access to on-premise database from Qlik Sense. https://cloud.google.com/vpn/
Many additional Google Cloud capabilities may also be used in relation to Qlik but are not directly discussed in this document. Much more can be seen at https://cloud.google.com/products/
Data Preparation – use Trifacta as a managed service within Google Cloud Platform to explore your data (CSV, JSON, or relational table formats) and prepare it for consumption. https://cloud.google.com/dataprep/
Several core and underlying infrastructure decisions need to be made and deployed before beginning the Qlik implementation, particularly the Network Tier where the Virtual Private Cloud is organized (address ranges chosen, connectivity with any on-premise datacenter established).
Connectivity between Qlik components, to Datasources and interaction with the Persistence Tier is highly dependent on how the customer has configured their Cloud Network (use of Public or only VPC-internal addresses, requirement for Proxy components, Firewall rules).
Is the Presentation or Web Tier going to be made publicly accessible via the Load Balancers using Public IP Addresses, or is access to the solution available only from the on-premise network and via additional end-user VPN Connectivity?
The customer must also give serious consideration to Identity and Access Management. How will users authenticate when accessing the Qlik solution? Does this require replication of Active Directory into a cloud component, or is there already a SAML Identity Provider that can be integrated with Qlik? Many other possibilities are also viable.
Readers are strongly advised to seek guidance from both Qlik Consulting and their account-specific GCP team to research options that are applicable within their specific IT & Security constraints. Some customers are commencing their transition to Cloud and making decisions that could have long-lasting implications or are negatively disruptive to adjust later.
The Design that is chosen may vary from the guidance in this document and should be clearly documented and agreed by all parties BEFORE commencing implementation.