post

Data Center War Stories talks to SAP’s Jürgen Burkhardt

by

And we’re back this week with the second instalment in our Data Center War Stories series (sponsored by Sentilla).

This second episode in the series is with Jürgen Burkhardt, Senior Director of Data Center Operations, at SAP‘s HQ in Walldorf, Germany. I love his reference to “the purple server” (watch the video, or see the transcript below!).

Here’s a transcript of our conversation:

Tom Raftery: Hi everyone welcome to GreenMonk TV. Today we are doing a special series called the DataCenter War Stories. This series is sponsored Sentilla and with me today I have Jürgen Burkhardt. Jürgen if I remember correctly your title is Director of DataCenter Operations for SAP is that correct?

Jürgen Burkhardt: Close. Since I am 45, I am Senior Director of DataCenter Operations yes.

Tom Raftery: So Jürgen can you give us some kind of size of the scale and function of your DataCenter?

Jürgen Burkhardt: All together we have nearly 10,000 square meters raised floor. We are running 18,000 physical servers and now more than 25,000 virtual servers out of this location. The main purpose is first of all to run the production systems of SAP. The usual stuff FI, BW, CRM et cetera, they are all support systems, so if you have ABAP on to the SAP in marketplace, you, our service marketplace, this system is running here in Waldorf Rot, whatever you see from sap.com is running here to a main extent. We are running the majority of all development systems here and all training — the majority of demo and consulting system worldwide at SAP.

We have more than 20 megawatt of computing power here. I mentioned the 10,000 square meters raised floor. We have 15 — more than 15 petabyte of usable central storage, back up volume of 350 terabyte a day and more than 13,000 terabyte in our back up library.

Tom Raftery: Can you tell me what are the top issues you come across day to day in running your DataCenter, what are the big ticket items?

Jürgen Burkhardt: So one of the biggest problems we clearly have is the topic of asset management and the whole logistic process. If you have so many new servers coming in, you clearly need very, very sophisticated process, which allows you to find what we call the Purple Server, where is it, where is the special server? What kind of — what it is used for? Who owns it? How long is it already used? Do we still need it and all that kind of questions is very important for us.

And this is also very important from an infrastructure perspective, so we have so many stuff out there, if we start moving servers between locations or if we try to consolidate racks, server rooms and whatsoever, it’s absolutely required for us to know exactly where something is, who owns it, what it is used for etcetera, etcetera. And this is really one of our major challenges we have currently.

Tom Raftery: Are there any particular stories that come to mind, things issues that you’ve hit on and you’ve had to scratch your head and you’ve resolved them, that you want to talk about?

Jürgen Burkhardt: I think most people have a problem with their cooling capacity. Even if we are — we are running a very big data center. We have a long capacity down the other side. There was room for improvement. So what we did is we implemented a cool aisle containment system by ourselves.

So there are solutions available, you can purchase from various companies. So what we did is, so first of all we measured our power and cooling capacity and consumption in very detail, and on basis of that we figured out a concept to do that by ourselves.

So the first important thing is today I think it’s standard. We have to change our requisitions, especially in the DataCenter which is ten years old, and which now also got the premium certificate. That data center, the rack positions were back front, back front, back front and we had thousands of servers in that data center.

So what we are now doing, already did to some extent in that data center is, we had to change the rack positions front, front to implement the cold aisle containment system. And we did — so IT did that together with facility management. So we had a big project running to move surplus shutdown, racks, turn whole — the racks in whole rows, front to front and then built together with some external companies, it was very, very normal easy method. Buying stock in the next super market more or less, build the containment systems and that increased where we have implemented it the cooling capacity by 20%.

Tom Raftery: Is there anything else you want to mention?

Jürgen Burkhardt: Within the last three to four years we crashed against every limit you can imagine from the various type of devices which are on the — available on the market, because of our growth in size. The main driver for our virtualization strategy is the low utilization of our development and training servers. So what we are currently implementing is more or less a corporate cloud.

When a few years ago, we had some cost saving measures, our board said, you know what, we have a nice idea, we shutdown everything, which has a utilization below 5% and we said well, that might not be a good idea, because in that case we have to shutdown everything, more or less. And the reason if you imagine a server and an SAP running, system running on it and a database for development purpose, maybe a few developers are logging in, this is from a CPU utilization, you hardly see it, you understand.
So the normal consumption of the database and the system itself are creating most of the load and the little bit development of the developers is really not worth mentioning. And even if they are sometimes running some test cases, it’s not really a lot. The same is true for training, during the training sessions there is a high load on the systems.

But on the other side these systems are utilized maybe 15% or 20% maximum, because the training starts on Monday, stops at — from 9:00 to 5:00. Some trainings even go only two or three days. So there is a very low utilization. And that was the main reason for us to say, we need virtualization, we need desperately and we achieved a lot of savings with that now and currently we are already live with our corporate cloud.

And we are now migrating more and more of our virtual systems and also the physical systems which are now migrated to virtualization into the corporate cloud. With a fully automated self service system and a lot of functionality which allows us to park systems, unpark systems, create masters and also the customers by himself. This is very interesting and this really gives us savings in the area of 50% to 60%.

Tom Raftery: Okay Jürgen that’s been fantastic, thanks a million for coming on the show.

post

Sentilla thinks of data centers, as data factories!

by

Data center

If you have been following this blog, you’ll know I have been profiling Data Center efficiency companies over the last few weeks. This week I take a look at Sentilla.

I talked to Sentilla’s CTO and co-founder, Joe Polastre, the other day and Joe told me that Sentilla came out of Berkeley where they had been looking at data analytics problems around large, potentially incomplete or inaccurate, streaming datasets. The challenge was how to turn that into a complete picture of what’s going on so people could make business decisions.

Sentilla takes an industrial manufacturing approach to Data Centers – in manufacturing you have power going in one side, and products and (often) waste heat coming out the other. In the same way in data centers you have power going in one side and coming out the other side you have the product (compute cycles) and waste heat. To optimise your data center you need to get the maximum data/compute (product) output with the minimum power in and the least waste heat generated. Sentilla thinks of data centers, as data factories!

Unlike most of the data center people I have talked to, Sentilla don’t talk so much about energy savings. Instead they emphasise maximising performance – getting the most out of your existing data centers, your existing storage, your existing servers, your existing power supply. By far the greatest saving from deploying Sentilla, Joe claimed, is not from the energy savings. That pales in comparison to the capital deferment savings gained from being able to delay the building of extra data center facilities by however many years, he said.

So how does Sentilla help?

Well Sentilla analyses the energy profile of every asset in the data center, whether metered or not, and makes recommendations to improve the planning and management of data center operations. I highlighted the “whether metered or not” bit because this is an important differentiator for Sentilla – they have developed and patented what they call “virtual meters”. These are algorithms which look at the work that a device is doing, and based on models which Sentilla have built up, and measurements they have done, as well as some benchmarks which are out there, Sentilla computes how much power is being used by that equipment.

The reason this is so important is because the most inefficient equipment in the data center is not the new stuff (which is likely to already be metered) but the legacy devices. These are the ones which need to be most carefully managed, and the ones where the greatest performance gains for the data center can be made. And because Sentilla can pull usage information from management databases like Tivoli, it means the Sentilla doesn’t need to poll every piece of equipment in the data center (with the increased network traffic and data that would generate).

Also, because Sentilla has its virtual meters, it is a software-only product and can therefore be rolled out very quickly.

The other nice feature Sentilla has is that it can identify the energy utilisation of virtualised servers. This is important because with the increasing ease of deployment of virtual servers, under-utilised VM’s and VM clutter are starting to become issues for data centers. VM clutter isn’t just an issue for energy reasons – there are also implications for software licensing, maintenance and SLA requirements.

I asked Joe about whether Sentilla is a SaaS product and he said that while they have a SaaS version of the product, so far most of Sentilla’s clients prefer to keep their data in-house and they haven’t gone for the SaaS option.

Finally I asked about pricing and Joe said that Sentilla is priced on a subscription basis and, apparently, it is priced such that for any modest sized data center, for every $1 you put into Sentilla, you get $2 back. Or put another way, Joe said, deploying Sentilla will generally mean that you reclaim around 18-20% of your power capacity.

Disclosure: Sentilla are a client (but this post is not part of their client engagement)

You should follow me on Twitter here

Photo credit The Planet

post

My “Green IT – driving efficiency, sustainability and enabling efficient working practices” presentation

by

Green IT – driving efficiency, sustainability and enabling efficient working practices

Conference organising company iQuest contacted me last year to ask me to deliver a keynote presentation at their Green IT Summit.

The event took place in Dublin yesterday and my keynote talk entitled “Green IT – driving efficiency, sustainability and enabling efficient working practices” is above.

The organisers prudently decided that they didn’t want to take the risk of any of their international speakers not making it to the event because of the ashcloud. This would have left them with a hole in the schedule at the last minute so they contracted the services of OnlineMeetingRooms and three of the presenters were able to present to the audience in Dublin, over an online video connection, without having to travel!

The title I was asked to present on was quite broad and I had 30 minutes to try cover it all so I had to go at quite a clip but the feedback has been extremely positive so it seemed to work out very well.