Upscaling Event Broker Services
You can upscale (or scale-up) existing
You may want to upscale your existing event broker service because you are exceeding the connections or queues provided by the limits of its current service class. Upscaling to a higher service class can also improve the performance of your event broker service.
For more information about upscaling, see the following:
For Dedicated Cluster customers, Solace cannot upscale an event broker service on the same day as a Kubernetes upgrade (or vice-versa). You must schedule upscaling of event broker services and Kubernetes cluster upgrades on separate days.
Considerations and Limitations for Upscaling Event Broker Services
Before upscaling your event broker service, review the following:
Upscaling Limitations
Upscaling an existing service has the following limitations:
-
You can't upscale from a Standalone or Enterprise-100 Standalone class to an HA service (nor scale from an HA to a Standalone or Enterprise-100 Standalone service).
-
You cannot upscale an event broker service deployed to a Customer-Controlled Container.
-
We only support upscaling event broker services deployed in Kubernetes environments that support volume expansion. Event broker services deployed using Static Local Provisioned volumes can't be upscaled. For more information, see Deployment Architecture for Kubernetes.
-
You can't downscale a service. For example, you can't downscale a service from Enterprise 10K to Enterprise 5K.
Account Considerations
Review the following account considerations:
-
Before choosing a service class to upscale to, consider whether that class meets your future growth requirements.
-
While event broker service configurations are retained after the upscale, you may need to change some configuration settings to take advantage of the new service class limits.
Considerations for Upscaling Event Broker Services in Customer-Controlled Clusters
If you deploy your event broker services in a Customer-Controlled Cluster, you must consider the following:
-
Both the existing and new resource requirements are necessary until the upscaling of the event broker service occurs. It's your responsibility to ensure the required infrastructure is in place to support the upscaled event broker service.
-
Solace upscales event broker services only in Kubernetes cluster versions we support.
-
Solace strongly recommends against upscaling event broker services on the same day as a Kubernetes cluster upgrade.
-
The Kubernetes cluster hosting the event broker service must support volume expansion. You must set the Storage Class parameter
allowVolumeExpansiontotrue.
To upscale an event broker service from Enterprise 250 to Enterprise 1K, you will require CPU, memory, and disk resources for both the existing Enterprise 250 event broker service and the new Enterprise 1K event broker service until the upscaling completes. We recommend that you contact your infrastructure team prior to booking your event broker service's upscale to ensure you meet these requirements. For more information, see General Resource Requirements for Kubernetes and Default Port Configuration.
After the upscaling of the event broker service completes, you can reallocate resources (CPU, memory, hard-drive) from the old event broker service class (that you upscaled from) if they aren't required or if you need to optimize the resources in your Kubernetes environment.
Solace recommends that you choose a service class size that accommodates not only future requirements (for example, the number of connections), but also aligns with your cost structure. Choosing a service class size that is too large with unused connections would not be cost-effective for your organization. Another consideration is that after upscaling the service class, you cannot downscale it. For more information on choosing the right service class, see Choosing Service Classes and Cloud Providers for Event Broker Services.
Upscaling an Event Broker Service
-
Before booking an upscale for your event broker service, ensure that your organization has the appropriate limits for the target event broker service class. You can check your account limits using the Cloud Console. If you don't have enough limits for the target service class, you can request additional limits using the Cloud Console.
-
Customer-Controlled Cluster customers must review the considerations and ensure your cluster meets the resource requirements for the target event broker service class when upscaling an event broker service.
Before you upscale your event broker service, ensure that you review the Considerations and Limitations for Upscaling Event Broker Services. You can typically book a time within one to two weeks to upscale your event broker service using Calendly. Solace recommends that a person with at least the Mission Control Manager role books the upscale, and we require a person with at least the Mission Control Manager role from your organization attend the upscale meeting.
To upscale an event broker service:
-
Plan a maintenance window for upscaling your event broker service that aligns with your organization's requirements and book a slot on Calendly.
-
After Solace contacts you to confirm a time, we upscale the event broker service at a meeting that someone from your organization must attend. For an overview of what occurs, see Overview of the Upscaling Process for Event Broker Services.
Planning a Maintenance Window to Upscale Your Event Broker Service
Solace recommends that you identify time periods to upscale your event broker service that work best for you and minimize any possible impacts on your client applications. You must select a time when someone from your organization with the Mission Control Manager role can join the upscale meeting.
During the upscaling procedure, the event broker service being upscaled can't use the message spool or hard disks.
HA event broker services are upscaled while in service. Your event broker service may experience a brief downtime of less than a minute when Messaging Connectivity switches messaging nodes during the upscale process.
Standalone event broker services should expect a Messaging Connectivity outage of up to 15 minutes during the upscale process. During this outage, the event broker service is unavailable to connecting client applications. The Messaging Connectivity outage of 15 minutes includes:
- the time required to upscale the event broker service
- the time to validate the event broker service
Overview of the Upscaling Process for Event Broker Services
After Solace confirms your upscale request and schedules a maintenance window, the following occurs:
-
Solace sends you a notification containing the date, time, and details of when your event broker service is to be upscaled. You must confirm that date and time, and the request details, and let us know if there are changes.
-
Solace sends a second notification to let you know that the upscaling of the event broker service will start during the agreed upon maintenance window.
-
During the maintenance window, which someone from your organization must attend, we upscale your event broker service to the requested service class.
-
After the upscale procedure completes (approximately 15 minutes), Solace sends a third notification to let you know the results of the upscale procedure. If the upscale procedure was:
Successful, proceed to the next step.
Not successful, Solace contacts you and works with you to correct the problem and to rebook a subsequent follow-up time to upscale the event broker service again.
-
After the upscale procedure, you can validate your event broker service's new class by selecting it in Cluster Manager and reviewing its information on the Configuration tab. The values under the following labels should reflect those your newly chosen service class:
Service Class
Endpoint (Queue)
Messaging Spool Size
If you previously increased the event broker services message spool size beyond the size of the newly selected class, the Messaging Spool Size shows the larger value. For more information, see Configuring Message Spool Sizes.
Max Queue Messages
Connections
Max Kafka Bridges
Max Kafka Connections
Max VPN Bridges
Maximum Message Size