Ironically when Dell, the company that built its success around supply chain management excellence, started the Data Center Solutions group to serve the “biggest of the big,” supply chain and procurement were just bit players.
I recently sat down with Chris Thompson who heads up the DCS supply chain and procurement organization and learned how this changed and the importance of his group in meeting the needs of this very unique customer set.
Some of the topics Chris tackles:
How his group helps customers get to revenue faster.
To what extent the DCS supply chain org is independent of Dell “normal” processes and procedures and to what extent it leverages them.
How has Chris’ group affected traditional Dell supply chain practices.
In my last entry I talked about how Steve Perlman, CEO and founder of OnLive joined the recent press round table we had in New York. OnLive is a cloud-based gaming company that launched earlier this year and whose servers were custom built by Dell’s Data Center Solution (DCS) group.
To give you a bit more insight into how the two companies worked together, here is a short video with Bruce Grove, OnLive’s director of strategic relations talking about the relationship between Dell and OnLive.
Some of the ground Bruce covers:
The value, as a start up, of working with someone who knows how to do supply chain, logistics and build tons of servers.
Working together as a team to design the servers (engineering teams on both sides as well as manufacturing teams).
Timothy Prickett Morgan of everyone’s favorite vulture-branded media site The Register attended a round table discussion we held a few weeks ago in New York. His piece from that event, which was focused around the cloud, was posted yesterday.
You should check out the whole article but here are some snippets to whet your appetite:
What DCS is all about
For the past several years – and some of them not particularly good ones – Dell’s Data Center Services (DCS) bespoke iron-making forge down in Round Rock, Texas, has been a particularly bright spot in the company’s enterprise business.
The unit has several hundred employees, who craft and build custom server kit for these picky Webby shops, where power and cooling issues actually matter more than raw performance. The high availability features necessary to keep applications running are in the software, so you can rip enterprise-class server features out of the boxes – they are like legs on a snake.
How we’re working with web-based gaming company OnLive
“These guys took a bet on Facebook early, and they benefited from that,” says Perlman [OnLive Founder and CEO]. “And now they are making a bet on us.”
OnLive allows gamers to play popular video games on their PCs remotely through a Web browser and soon on their TVs with a special (and cheap) HDMI and network adapter. The games are actually running back in OnLive’s data centers, and the secret sauce that Perlman has been working on to make console games work over the Internet and inside of a Web browser is what he called “error concealment”.
DCS had to create a custom server to integrate their video compression board into the machine, as well as pack in some high-end graphics cards to drive the games. Power and cooling are big issues. And no, you can’t see the servers. It’s a secret.
Last week a couple of us went down to San Antonio to help represent the OpenStack project at Rackspace’s partner summit. While there I met up with the VAR Guy. Mr. Guy got me chatting about Dell’s Data Center Solutions group, where we’ve been and where we’re going. Below is the resulting video he put together featuring myself and San Antonio’s greenery. (See the original article this came from).
Some of topics I tackle:
How Dell’s Data Center Solutions Group is designing servers for high-end cloud computing
How Dell is integrating hardware with software in cloud servers
Coming soon: Dell Cloud Solution for Web Applications/Leveraging Joyent‘s software
One of the key ingredients for the success of any open source project is a strong community manager. Coming on board to fill that role for the not-quite three-month-old OpenStack project is Stephen Spector. (If you’re not familiar with OpenStack, it’s an open source cloud platform).
Stephen made his first public appearance in his new role today at the Rackspace partner summit in San Antonio. I was able to catch Stephen first thing this morning before the summit kicked off.
Some of the ground Stephen covers:
His background: 14 yrs at Citrix. He initially ran developer alliance programs. He spent the last 3yrs running the Xen.org community.
Why Stephen joined OpenStack (he jumped at the chance to build a community from scratch).
He sees his role as that of a communication conduit
One of his first tasks is to find out who makes up the community e.g. developers, users, students, research, partners..
He’s very interested in making events like next months design summit successful as well as the importance of globalization.
A couple of days ago Bret Piatt, who handles Technical Alliances for OpenStack, came up to Austin to have further discussion with our team’s software engineers around OpenStack. If you’re not familiar with OpenStack, it is an open source cloud platform founded on contributed code from Rackspace and NASA’s Nebula cloud.
The project was kicked off a couple of months ago at an inaugural design summit held here in Austin. The summit drew over 25 companies from around the world, including Dell, to give input on the project and collectively map out the design for the project’s two main efforts, Cloud Compute and Object Storage.
Since the summit, and the project’s subsequent announcement the following week at the OSCON Cloud Summit, the community has been digging in. The first object storage code release will be available this month and the initial compute release, dubbed the “Austin” release, is slated for October 21. Additionally, the second OpenStack Design Summit has been set for November 9-12 in San Antonio, Texas, and is open to the public.
OpenStack visits Dell
During Bret’s visit to Dell he met with a bunch of folks including two of our software architects, Greg Althaus and Rob Hirschfeld. The three talked about how things were going with the project since the summit as well as specific ways in which Dell can contribute to the OpenStack project.
Below you can see where I crashed the three’s whiteboard session and made them tell me what they were doing. I then followed them, camera in hand, down to the lab where Greg and Rob showed Bret the system that we have targeted for running OpenStack.
Some of the topics (L -> R) Bret, Greg and Rob touch on:
Bret: Getting ready for the object storage release in September and compute in October. Looking to get the right hardware spec’d out so that people can start using the solution once its released.
Rob: Learning about how the project is coming together since the design summit. Interested in how the 3 code lines, storage, NASA compute and Rackspace compute, along with the input that was gathered at the Design summit and community input, are coming together.
Greg and Rob take Brett to the lab to show him the C6100 which could be a good candidate for open stack.
Next step, getting OpenStack in the lab and start playing with it.
iland is a provider of cloud computing infrastructure with high-availability data centers specifically designed for cloud computing in Boston, Washington D.C., Houston, Los Angeles, Dallas, and London. To stay competitive in the cloud infrastructure business, iland needs to gain as much value as possible from every watt of power and every square foot of data center floor space.
One day over lunch they were introduced to the PowerEdge C6105 and were impressed with the product’s density and power efficiency. One thing led to another and here is iland’s CTO Justin Giardina talking about why they are so interested int the PowerEdge C6105.
Some of the points Justin makes:
The primary reasons iland wanted to talk to Dell about the 6105 were density and power draw.
iland can stack 20 servers in one cabinet and since each 6105 has 4 servers in it by filling the cabinet with 6105s they can in affect get 80 servers (4X the compute power per cabinet).
The system only pulls 3 amps from both power supplies.
A couple of weeks ago a group from salesforce.com paid a visit to Dell. Among other things, they came to discuss their new product “Chatter” that Dell has recently launched internally and who’s virtues Michael Dell has tweeted. Among the salesforce crew was Sean Whiteley, VP of product marketing. I was able to get some time between meetings with Sean and learn more about Chatter.
Some of the topics Sean tackles:
How Chatter has done since its launch on June 22. What type of traction they’ve seen with customers.
How Chatter differs from other internal social media platforms (hint: not only can you follow people; records, objects and information within your business applications have feeds as well, e.g. your notified when a presentation changes or a sales deal you’re following moves to a different stage.)
How the idea of Chatter came up. What role chairman Marc Benioff and his use of Facebook played.
Currently Chatter is tied closely to CRM but it will be tied to other apps going forward.
They believe that many more folks will use Chatter than usesalesforce.com.
J.P. Rangaswamy is British Telecom’s chief scientist and a very interesting fellow. At the Cloud Summit at OSCON last month he delivered a talk on the future of the cloud. I was quite intrigued so I grabbed him during the break to learn a bit more about a few of the concepts he presented.
Some of the topics that J.P. tackles:
Many of the best utilities we’ve built: Internet, Web, wireless environment etc are built on fundamentally frail best-effort infrastructures.
In order to gain predictability you sacrifice a lot of the original value eg. QWERTY
He’s not against SLAs, J.P.’s against the throwing away of value under the guise of false beliefs in SLAs.
What is the key area in the cloud that needs to be shored up? Interoperability. Security is overplayed, just look at the development of the Web.
Need to concentrate on federation as a mind set — the ability to create services that are daisy chaining select pieces from a variety of each, vs integrated vertical stacks.
He’s worried about SLAs because the things people are doing to stop SLAs from being light weight are actually things that prevent interoperability.
The week before last our team decided to divide and conquer to cover two simultaneous events. Half of us headed to Portland, Oregon for OSCON and the other half stayed here in Austin to participate in hostingcon.
The Hostingcon keynote
Among those participating in hostingcon was my boss Andy Rhodes who gave the keynote on Tuesday. Here’s are the slides Andy delivered:
(If the presentation doesn’t appear above, click here to view it.)
The idea of the keynote was to share with hosters the five major lessons we have learned over the last several years working with a unique set of customers operating at hyperscale. Those five lessons are:
TCO models are not one size fits all. Build a unique model that represents your specific environment and make sure that you get every dollar of cost in there. Additionally, make sure that your model is flexible enough to accommodate new info and market changes.
Don’t let the status quo hold you back. Not adapting soon enough and delays in rolling out solutions can cost you dearly.
The most expensive server/storage node is the one that isn’t used (sits idle for 6-12 weeks) or the one you don’t have when you need it most.
Don’t let Bad Code dictate your hardware architecture.
Don’t waste time on “Cloud Washing.” Talk to your customers about real pain points and how to solve them.
So, it seems that cloud best practices will help companies reduce their physical infrastructure, which seems to be a bit counter-intuitive, given that Rhodes is representing a hardware provider. But it makes sense. Given the never-ending list of projects for IT staff, and as they drive down costs, their business will grow, and they’ll be able to increase their IT spend for innovative efforts. “What we’re hoping to do is let you do more with less.”
At OSCON last week I ran into a compadre from a previous life, Fred Kohout. Fred is now the CMO at UC4, a pure play software automation company, and he, like I, was in Portland to attend OSCON and the Cloud Summit.
At the summit Fred did to me what I’ve done to so many others, he got me on the receiving end of a video camera to talk about where Dell plays in the cloud and how we see the cloud evolving.
You can check out Fred’s blog from the Summit where he posted my video as well as the interview he did with another former compadre, Peder Ulander, CMO at cloud.com.
Don’t touch that dial
If you’re interested in OSCON be sure to stay tuned. I’ve got four more interviews from the event that I will be posting soon.
At the inaugural design summit for OpenStack, an open source set of technologies for building clouds, Nebula’s chief architect Josh McKenty played a prominent role in leading the assembled folks. I caught Josh during a break and chatted with him about Nebula and NASA’s role in the newly announced OpenStack project. Here’s what he had to say:
Some of the topics Josh tackles:
What is Nebula (hint: NASA’s, primarily IaaS, cloud computing platform)
The history of Nebula and how it morphed from nasa.net.
Why NASA wants a cloud – and the importance of having an elastic set of resources.
NASA and Nebula’s use of open source and how it has evolved (they don’t simply fling tarballs over the wall anymore and they can use licenses other than the “NASA open source agreement”)
A match made in heaven: NASA has put together a strong compute platform and was looking to building a real object store, Rackspace had a strong object store and work looking for a new compute platform.
Today at E3, OnLive Inc is kicking off the roll out of its cloud gaming service. OnLive, whose motto is “Just Play,” leverages broadband and the cloud to deliver on-demand gaming titles directly to users’ PCs, Macs or even TVs.
Square Enix's Batman: Arkham Asylum -- one of the first batch of games available from OnLive
This new service could prove to be a real “game changer.” As Dell Data Center Solutions director Andy Rhodes, helping with the launch at E3 explains, “I see it as the start as of a move of processing power from consoles into data centers…from the center of the living room into the data center.”
Building the OnLive Cloud
So what’s behind this gaming cloud, Dell of course (well, at least a good part of it). The Dell Data Center Solutions (DCS) group began working with OnLive a few years back to design and build custom-tailored systems for the OnLive platform.
The problem statement for the solution was to create an infrastructure that supported the streaming of HD-quality video game over the internet, drove down the total cost of ownership and allowed OnLive to scale quickly as the company grows. The DCS team worked directly with the folks from OnLive to architect an ultra-dense and uber-power efficient infrastructure solution designed around OnLive’s super secret hardware components and software. Thousand of these customized systems are now deployed at OnLive data centers around the country.
Plug and Play Racks
By leveraging the DCS supply chain and fulfillment chops, Dell is able to deliver pre-integrated fully racked solutions that can be hooked up and powered on within hours of arriving at an OnLive data center. Going forward Dell will continue to work with OnLive to create new infrastructure architectures for future generations of the service.
Game on! (and on, and on and on)
Electronic Arts' Mass Attack 2: available via OnLive
Who’s on First?
The initial batch of 23 titles available to OnLive subscribers include:
Over the past week I have presented Dell’s thoughts and capabilities around cloud computing to several different groups from the U.S. military. In preparation for these talks I did some research into what’s happening in the wild and wonderful world of federal cloud computing. Here are a couple things that I found particularly interesting:
Psych!
In the past I have used the General Services Administration’s cloud RFQ issued last July, as an example of how the government is boldly sallying forth into the cloud. Turns out that in February they withdrew the RFQ saying basically that too much had changed since the RFQ was issued and that they need to regroup and get a solid view of the customer and market landscape before writing a new one.
Speaking of snags, Apps.gov which was launched last September as “an online technology supermarket for federal agencies” has not been the success that Federal CIO Vivek Kundra had hoped for. According to the WSJ, “concerns about compliance with security requirements and terms of service have prompted many agencies to bypass Apps.gov.”
But wait, there’s more
The above being said, the US government has a ton of cloud projects its working on. To get smart on the littany of efforts, check out the State of Public Sector Cloud Computing report that Vivek Kundra issued last week.
Stay tuned for my next entry that will talk about how the Military is “forging” ahead.
When I was out in the Bay Area for our launch I stopped by data warehouse and analytics player Greenplum. Greenplum is one of the first three members in our Cloud Partner program (the other two are Canonical and Aster Data.) I sat down with Greenplum’s President and founder Scott Yara to talk about the company and where they’re going:
Some to the topics Scott tackles:
Whats happening in the world of data.
How Greenplum began with the open source PostgreSQL database platform and over the last 7-8 years have refactored it and built a massively parallel database kernel engine.
How it works: Greenplum takes the data and physically distributes it across all the Database segments and operates on the data in parallel. This parallel approach allows Greenplum to process data 10-100x faster than conventional databases.
Who is using it: Skype, Fox Interactive, NTT docomo, Deutsche Bank, retailers, large healthcare companies.
The enterprise data cloud initiative – Setting a new type of analytics infrastructure that takes advantage of virtualization and the latest in general purpose and multi-core systems and is centered around self-service principles.
While a lot of folks are excited about writing apps to the iPhone, the platform that Scott and crew gest really excited about writing to are 2 socket Nehalem servers with a bunch of disk drives behind them.
How someone would go about getting started with Greenplum.
Last month when I was out in the Bay Area for our launch, I stopped by the offices of salesforce.com. I visited with some folks that I used to work with in a past life and then grabbed some time with Salesforce’s VP of product marketing, Sean Whiteley.
Here is what Sean had to say:
Some of the topics Sean tackles.
The idea behind salesforce.com (SFDC): In 1999 founders Marc Benioff and Parker Harris looked at Amazon and wondered why businesses couldn’t manage and get insight into their customers with the same ease as they interact with their favorite website.
Given that SFDC is built on a model of “multitenancy” how do they address security concerns when they are brought up.
Force.com: what it is and how it came about. Also the advent of AppExchange, where you can shop for applications that let you extend the cloud applications that you use to run your business.
What salesforce.com and Dell are doing together to address small and medium businesses: providing a business in a box, helping organizations focus on their core business rather than IT.
Last but not least in our tour of the first flight of servers in Dell’s PowerEdge C line is the C2100. This was filmed last month at our launch right after we closed down our whisper suite. Click below and join solutions architect Rafael Zamora as he leads you through the thrills, chills and spills of this cool new machine.
Spoiler Alert! A few Highlights
This machine is a great high performance data analytics and cloud optimized storage server. It’s perfect for use in conjunction with software from our Cloud ISV partners Joyent, Aster Data, Canonical or Greenplum.
The entire front can be jam-packed with disk drives for mega storage. You can get up to 24TB by loading twelve 3.5inch 2TB drives.
While the C2100 has same system board as the C1100, it comes with twice the real estate in a 2U form factor.
There’s even a platform for two additional drives that you can use for a bunch of different purposes such as separating production and non-production traffic.
As part of the CIO series we are doing with Information Week, I did a webinar on Wednesday called The Journey to the Cloud. If you want to get a feel for some of the big issues and ideas addressed during the webinar, check out moderator Fred Paul’s entertaining blog entry. Here is the deck I used:
If you’re interested in listening to the webinar you can get access to it here (you will need to register). I spent about 20 minutes on the presentation and 30 minutes fielding questions from the audience. One question submitted that we didn’t have time to answer was why Fred Paul and I both have first names for last names. That will have to wait for another webinar.
Our first Cloud white paper is now available and I’m really happy with the way it turned out. Being relatively new to Dell I didn’t know if I would be “compelled” to mention product in it or not but I’m happy to say that the paper focuses solely on the trends behind, and characteristics, of cloud computing . I want to give a big shout out to Intel who helped to fund this and also didn’t insist that we mention their products.
This first paper is a short and basic introduction to cloud computing. We are working on a follow-up that will pick up where this leaves off and dives deeper. Stay tuned.
If you’re wondering about the funky game show-like setting, I shot this after hours on the day of our launch in the whisper suite. Your guide, as before, is the incomparable Dell Solutions Architect, Rafael Zamora.
A few highlights
The C1100 is a high memory, cluster optimized, compute node
Dont let its slim pizza box looks fool you, upfront you can pack either four 3.5 inch drives or ten 2.5 inch drives.
For high memory optimized compute you can get 18 DIMM sticks for 144GB of RAM.
Comes with your choice of either Intel’s Nehalem or Westmere processors.
Raf also gives a couple of examples of recent customers and how they’ve decide to configure their units.
Posted by Barton George 
(well, at least a good part of it). The Dell Data Center Solutions (DCS) group began working with OnLive a few years back to design and build custom-tailored systems for the OnLive platform.
