With the 1.5 release of Portworx, we have drastically simplified how to install Portworx on…
April 27, 2021
Automatically optimize the performance of your Kubernetes applications
One of the benefits of the Portworx data management platform is the flexibility it provides for all stateful workflow types. WordPress blogs, search databases, streaming services and message queues can all be configured at a per workload, per container, and per volume level. In this blog, we will explore a feature from the recent 2.7.0 Portworx release. As of this release, there is a feature that allows you to let Portworx “auto-magically” configure the best IO profile for your containerized workload. So, what does this actually mean?
Portworx IO profiles let you optimize the performance of your Portworx volumes by matching the type of workload you’re running with a suitable IO profile. IO profiles change the way a Portworx volume interacts with the underlying storage disks to improve IO traffic for different workloads.
How does the “auto” profile benefit you?
Prior to the auto profile storage, admins, operators, and users that create storage classes would need to understand the workloads being served by such storage classes before they started provisioning storage from a given storage class. With the auto profile, Portworx will attempt to automatically apply an IO profile that is most appropriate for the data patterns it sees, even after dynamic provisioning has occurred. It does this by continuously analyzing the IO pattern of traffic in the background. Ultimately, this helps simplify StorageClass design and allows users to consume similar storage classes across various service types of workloads with variable IO patterns.
In Diagram 1, the left side of the diagram shows how you would provide storage classes to workloads with different types of IO prior to using the auto profile. On the right side, you can see that using the auto profile allows you to reduce this complexity and use similar parameters while allowing Portworx to detect and apply the IO profile during runtime.
Check out the auto profile in action in the below demonstration.