Systems and methods of auto-scaling a virtual desktop environment
A system for optimizing OS disk resource utilization in a dynamically auto-scaling virtual desktop environment includes: a scalable virtual desktop environment in which a VM is selectively powered on and off, wherein the VM is associated with a corresponding OS disk that is alternatively assigned to a higher performance OS disk tier or a lower cost OS disk tier; a server including a controller controlling the powering on and the powering off of the VM and assigning the OS disk tier to the VM; and a memory coupled to the controller, wherein the memory stores executable code programming the controller to: in response to the controller powering off the VM, assign the lower cost OS disk tier to the virtual machine; and prior to the controller powering on the virtual machine, assign the higher performance OS disk tier to the virtual machine.
1 . A system for dynamically auto-scaling allocated capacity of a virtual desktop environment comprising:
a scalable virtual desktop environment comprising a base capacity resources, including base compute resources and base storage resources, and a burst capacity resources, including burst compute resources and burst storage resources;
a server including a controller controlling the operation of the base capacity resources and the burst capacity resources;
a memory coupled to the controller, wherein the memory is configured to store program instructions executable by the controller;
wherein, in response to executing the program instructions, the controller is configured to:
provide a host pool sizing tool including controls for a base host pool capacity, a minimum active host capacity, and a burst beyond base capacity;
provide an available sessions scaling out logic tool;
in response to receiving a log in request from a first user device, connect the first user device to a first host pool to which the first device user is assigned;
in response to receiving a log in request from a second user device, connect a second user device to a second host pool to which the second device user is assigned;
execute a load-balancing module to determine a first session host virtual machine to which to connect the first user device;
execute an auto-scaling module comprising an auto-scaling trigger and a conditional auto-scaling action, wherein, in response to recognition of the conditional auto-scaling action, the controller powers on or powers off one or more base capacity resources or creates or destroys one or more burst capacity resources;
in response to the controller powering off one or more base capacity resources, assigning a lower cost OS disk tier to each of the one or more base capacity resources that have been powered off; and
prior to the controller powering on one or more base capacity resources, assigning a higher performance OS disk tier to each of the one or more base capacity resources that have been powered on.
2 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide a pre-stage hosts tool defining one or more work days, a start of work hours, and a number of hosts to be active by the start of work hours.
3 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide a CPU usage scaling out logic tool including an auto-scale trigger and a number of hosts to scale out when the CPU utilization exceeds a given capacity for a given duration.
4 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide an average active sessions per host scaling out logic tool including an auto-scale trigger and a number of hosts to scale out when the average active sessions per host exceeds a given number.
5 . The system of claim 1 , wherein, the available sessions scaling out logic tool includes an auto-scale trigger selection, a maximum sessions per host, and a minimum number of available sessions that need to be maintained at all times.
6 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide a working hours scaling out logic tool based on one or more of a day of the week, a time range, a setting for power off aggressiveness, and a setting for power on timing.
7 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide a CPU usage scaling logic tool including scale out and scale in definitions including an auto-scale trigger and a number of hosts to scale out when the CPU utilization exceeds a given capacity for a given duration, and a number of hosts to scale in when the CPU utilization drops below a given capacity for a given duration.
8 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide an average active sessions scaling logic tool including scale out and scale in definitions including an auto-scale trigger and a number of host to scale out when the average active sessions across all hosts exceeds a given value, and a number of hosts to scale in when an average active sessions across all hosts is below a given value.
9 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide a scale in restrictions tool including times in which scale in will be restricted and a level of scale in aggressiveness to apply.
10 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide a scale in messaging tool including when to send a message relative to a scale in action.
11 . The system of claim 1 , wherein, in response to executing the program instructions, the controller is configured to provide an auto-heal broken hosts tool including instructions defining how to identify when to heal a broken session host, including applicable WVD agent statuses and whether hosts with any session or with active sessions should be acted upon, a number of restart attempts, a time between restart attempts, and how to handle unrecoverable hosts.