In today’s “Acronym Soup” blog post, we once again step over on the manageability side of things and ask the question: What is PSME?

PSME is an acronym for “Pooled System Management Engine”, which gives us a hint about its actual meaning. Just by looking at the words standing behind this acronym, we can make a guess that PSME is a kind of technology that enables the management of “pooled” or grouped computing systems. So let’s now examine what this technology is even further, by first taking another look at Intel® Rack Scale Design Architecture.

An easy way to understand the Pooled System Management Engine, or PSME, is to know that it is a key component of Intel® Rack Scale Design Architecture, or RSD for short, which we have discussed in previous posts. According to Intel’s RSD website, Intel RSD allows users to “dynamically compose resources based on workload-specific demands… Intel® RSD complements virtualization, cloud computing, and other software-oriented approaches by dynamically composing workload-optimized hardware while at the same time allowing workloads to run on bare-metal and do it all with a single management console.”

Why is “pooling” of compute resources so important and useful?

The above quote from Intel gives us a much better idea of what pooling is, and why it might be an important and useful technology. It was developed for use in cloud computing data centers, where a large number of servers are in operation. The ability to select and aggregate, on-demand, only the computing resources required for a specific workload or computing task is exciting and transformative. It opens the doors for improved efficiency in terms of hardware inventory, power consumption, compute performance, and many more benefits. Fundamental to this technology is the ability to see the resources independent of the physical hardware on which they may reside in the data center; this separation is key to the ability to pool/aggregate them in specific combinations for specific tasks.

How is PSME and pooling relevant to AMI technology and products?

Several of our MegaRAC manageability solutions are designed to integrate into the Intel RSD ecosystem. First among them is MegaRAC Composer®, which was developed by AMI to be fully compliant with the Intel® Rack Scale Design and DMTF Redfish™ standards for server and resource management. MegaRAC Composer is best described as pod management software; the “pods” in this usage which are managed by MegaRAC Composer are those same disparate compute, storage and power resources that might exist in a given data center. MegaRAC Composer is a single management portal, like a single pane of glass through which users can browse and select physical resources at the rack, chassis and system level.

Once the workload is identified and the right resources are selected, administrators can then compose and assign those physical resources to create a logical node, which provides the advantage of demand-driven dynamic scaling to optimize datacenter resource utilization. MegaRAC Composer also allows for the composition of physical resources based on templates, which can then be stored in the software and reused as a time-saving feature. In addition, MegaRAC Composer gives users the ability to power on, force off, and gracefully shut down composed nodes.

Thanks for reading today’s Tech Blog, we hope you found this bit of Acronym Soup refreshing and enjoyable! Feel free to drop us a line via social media or our Contact Us form and let us know – and what you might like to see in future posts!