The cost and operations of a datacenter are affected by many different factors which, until recently, did not impact business competitiveness significantly enough to matter. Hardware utilization, as long as it was twice as good as a few years ago, was acceptable. Software patching, as long as it didn't break anything, was acceptable. The cost of cooling, the cost of facilities, the efficiency of operations, as long as they were not too bad, as long as they let the business make a profit, were acceptable.

In a digital economy, where the cloud is the new server, acceptable is no longer good enough. In cloud, or what we call digital, economies, every penny counts. Every second counts. The competition is too fierce, the pace is too fast to ignore anything. Everything that's happening in your datacenter counts.

Video: Ericsson Datacenter Automation Platform (DAP)

This video recorded at Intel® Developer Forum in San Francisco this August, shows you some of the ways Ericsson Datacenter Automation Platform (DAP) aggregates, into easy to understand dashboards, the types of data that impact datacenter cost and agility. Narrated by James Malachowski, Director of Product Management at Ericsson.

 

 

If you're interested in exploring our ideas in more depth, we recently published an e-book called "Transform Your Datacenter Into Your Competitive Edge." 

Download the eBook

Transcript of DAP Demo

Speaking: James Malachowski, Director of Product Management at Ericsson.

Ericsson-hyperscale-cloud-blog-infrastructure-IDF-datacenter-automation.jpg

The Ericsson datacenter automation platform is designed to aggregate all the

disparate data sources from all the machines across your entire datacenter.

We've prepared for the demos a couple of aggregate dashboards showing some really interesting insights captured from a live customer environment.

What I'm showing you here is an aggregate of CPU metrics across an entire set of HP machines across the entire datacenter, giving you a histogram all the utilization to show you how much capacity do I have. As you can see, the hardware in this datacenter is not heavily utilized.

We've also built some dashboards to give you aggregate information about things like power and temperature data. How do we start to determine things like average power consumption at idle. How do we start to figure out the ambient temperature in the datacenter. We can give you all that data and more.

So drilling a little further, one of the interesting things we were able to showcase with this customer was issues with firmware. You can see in this environment that I've made a few dashboards related to the BMC, the NICS, the BIOS, the disks. And we can see see a very fragmented, very complex environment. Firmware can cause issues. It can cause security vulnerabilities. These are the types of things we want to keep up to date. And want to be able to use the DAP to to then automate or mediate those systems as we see fit.

We've also built an aggregate dashboard showing you all the disparate aggregate amounts of components. You can see in this environment we have systems from a wide variety of vendors. We've got disks from a wide variety of vendors. We've got networking devices we're pulling data from. Again, all this is designed to give the operator that 10,000 foot view of the datacenter.

What if I wanted to go back in time? Solve a particular problem? Understand at the micro level what's happending. The Ericsson DAP has a unique way of storing data. We treat all data in the datacenter as time series, which allows us to go back in time and look for interesting insights and find problems. And what we found here is the ability to look at the changes in the machine over the last several months. As you can see, this machine has been changed many times. I can zoom in on this timeline and look at various aspects of this system. We can drill down and see that the networking was dramatically reconfigured on the system. We've installed several new software packages and the like.

As an operator, I can take this data to make decisions, drive insights. Where we're going with this is that Ericsson is going to be building a variety of data models, machine learning algorithms to drill into this data and start to predict what is going to happen, and be proactive with notifications.

The E DAP is designed to give you the full visibility and transparency into the dynamc landscape of a changing datacneter environment.

In the future we are going to take this data and applying advanced machine learning techniques to be more predictive and proactive in helping you understand the complex dynamics of a changing landscape, from the lowest levels down at the physical layer all the way up to the application.

If you're struggling to keep track of your datacenter and its complex aspects of a changing environment, we can help.

For more information

Background photo by Deirdre Straughan.


Cloud Infrastructure

Rick Ramsey

I started my high tech training as an avionics technician in the US Air Force. While studying Economics at UC Berkeley, I wrote reference manuals and developer guides for two artificial intelligence languages, ART and SYNTEL. At Sun Microsystems I wrote about hardware, software, and toolkits for developers and sysadmins, and published “All About Administering NIS+.” I served as information architect before joining BigAdmin, which morphed into the Systems Community at Oracle. I left Oracle in May of 2015, and now write for Ericsson.

Follow Rick Ramsey:
Rick Ramsey

Discussions