Prior to the Intel Xeon X5500 Server Platforms*, measuring server power was done via expensive equipment and could only be performed in a discrete fashion.  Unless you had tons of monitoring equipment to mash-up your power data - it was a tedious process.  Now, using Intel DCM and Node Manager - you can pull multiple servers worth of power info to make some important power decisions in your datacenter.

First of all, you need to baseline your workload.  If you're confident that you can replicate workload patterns then you've got a starting point.  Otherwise, it's usually a good idea to start monitoring and looking for some cyclical patterns and/or common data points (time, power, thermals, etc) to keep track of.

In this scenario (like in my last blog) we're using a SQL workload which can be modified to run the CPU at high levels for a relatively set amount of time.  The base workload runs for 7 min 30 seconds, as shown in the Intel DCM screencap below.

In this test case: Idle power for the 4 servers is 782W, and under load - the power increases to 1174W - which is a delta of 392W.  This power increase occurs when work is given to the server and the P/T states react to the workload and increase power/voltage to the system to increase performance.  Exactly what we've been used to seeing even since EIST was introduced several years ago.

Now, what I'll show you is something that may be very interesting in scale... I will power cap the servers by 20W each, and set the Intel DCM Power Policy to only allow 1095W for the 4 servers in the rack.

What is awesome here is that we can still finish the workload in the same 7 minutes 30 seconds.  So essentially, we have saved 80W of power for each set of 4 servers and still get the same amount of work completed!  In a large datacenter this can be HUGE in energy savings.

Let's do some quick math:  20W power savings per serer x 10,000 servers = 20kW power savings and you still get the work done.  I hope I just helped some of you server admins get some new ideas on your next "I need a raise" talk with your manager

*your mileage may vary, so test your own workloads and report out!

One of the recurring themes that I've been noticing from end-users who are testing or evaluating Intel Intelligent Power Node Manager (or Node Manager) - the question is "How do we turn it on or off?"  To put it simply - when you have a Node Manager capable platform - you can simply put it to work and let your power policies decide when to enable/disable the features...

So let me step things back a bit and talk about the technology itself first.  Node Manager is very much like any *T technology that Intel has deployed over the past several years, it's an ingredient - or in this scenario a mix of ingredients that is available at the platform level.  Here are the 'ingredients' that when combined, give you the ability to monitor/manage power, and in some cases monitor thermal events.

• The platform is based on the Xeon 5500 Series Chipset (codename Tylersburg-EP) server board
• Xeon 5500 Series Processors (codename Nehalem EP)
• Node Manager Enabled Firmware with the Manageability Engine
• Server chassis components that meet IPMI 2.0 specifications for monitoring (e.g. thermal monitoring)
• PMBUS Power Supply - this communicates with the Baseboard Management Controller (BMC) for platform power usage

For those of you wanting to get your hands on this technology TODAY - check out the Intel Server linueup:

• Intel® Server Board S5500WB (codename Willowbrook) which is optimized for IPDC deployment, and supports IPMI 2.0, Intel Intelligent Power Node Manager, and can also support the Data Center Manageability Interface (DCMI) 1.0 specification.
• Intel® Server Board S5520UR (codename Urbanna) is the mainstream Enterprise platform which support IPMI 2.0 and Intel Intelligent Power Node Manager

Both platforms work in conjunction with Intel® Data Center Manager (Intel® DCM) which is the SDK which provides power and thermal monitoring and management.  This SDK allows group and policy based management for single server, rack, logical group, lab, or whole datacenter models.

Ok - so that reads like a bunch of marketing stuff... but here's the 'guts' of the technology...

When you purchase a Node Manager enabled server, there are a few simple steps to take to set things up to monitor/manage your server.

Most likely you'll need to setup your BMC, Intel provides a CD based implementation to help with this in our servers - it's called the Intel Deployment Assistant.  This lightweight OS bootable CD can setup the most common BIOS settings, check versions of firmware and update them via Internet connection to ensure you have the latest BIOS, BMC, ME and Sensor firmware.  Each OEM will have their own methods but should be similar in function when it comes to setting up the server for monitoring.

The BMC needs an ip address, netmask, and default gateway setup - and according to IPMI specifications - you can also set the administrative (user) access rights if you would like to tighten down security a bit.  Once you have these access points setup - you can utilize standard IPMI commands to communicate with your server or use Intel DCM to really  'visualize' the capabiliites of Node Manager.

Here's a great demo video showcasing some of the Node Manager & Intel DCM use cases:

How many of you have worked with IPMI management before?

The technology that has been around for a while, but now Intel has put automation and policy based management features into the platform - thereby reducing costs, increasing responsiveness to power policies, and also making Xeon Servers more energy efficient than before.  Many of our customers are asking for Node Manager enabled servers - is your OEM on track to deliver?