When exploring the io.Connect Desktop processes shown in Task Manager, a customer noticed several Web Group App and Workspaces App process running at the same time, even when only a few windows were open and wondered why. This article explains that behavior, it’s design, why it was implemented this way and how to configure it. To keep grouping and Workspace operations responsive, the platform pre-creates and caches a small pool of these instances so they are ready to use the moment they are needed, rather than being spun up on demand.
Why do I see more than one Web Group App process?
To improve group performance and the user experience, the platform pre-creates and caches a small pool of Web Group App instances so they are ready to use the moment they are needed, rather than being spun up on demand. The default pool size is 3, configured with the "pool" property. Beyond the cached pool, launched additional web groups in use will run as its own Web Group App instance.
See the "pool" property in the application configuration documentation: io.Connect Desktop Documentation - Configuration > Application
Why do I see more than one Workspaces App process?
The same pooling behavior applies to Workspaces. The Workspaces App uses a pool of pre-cached, semi-initialized Frame instances, configured with the "framePool" property (3 by default). A higher number increases memory consumption; a lower number increases the likelihood of delays during Workspace operations. When more Frame instances are needed than are available in the pool, additional Workspaces App instances are created and consume additional resources. Note that multiple Workspaces opened as tabs in the same Frame do not each require a separate Workspaces App process.
See the "framePool" property in the Workspaces documentation: io.Connect Desktop Documentation - Windows > Workspaces > Overview
Performance implications
A higher number keeps more instances pre-cached and ready, which improves responsiveness when opening new groups or Workspaces, but increases overall RAM usage. A lower number reduces overall RAM usage, but opening a new group or Workspace may become slightly slower, since an instance has to be created on demand. This lets platform administrators find the best balance between memory footprint and performance for their specific needs and requirements.