We’ve been watching the ARM market develop over the past few years as these highly efficient chips that have been driving tablets and cell phones have been finding their way more and more into hyperscale servers. Well watch no more, today were are sallying forth. Why now? Because some of our biggest customers have told us that they felt the time is now to start working with these low powered, highly efficient chips for their servers.
HW + SW = Solution
Today we announced that we will be shipping the new Dell “Copper” ARM servers via a seed unit program to select hyperscale customers worldwide. But a server does not an ecosystem make so we are doing what we can to help partners and developers get started building out applications for the platform. Given that two of the key areas where the extreme efficiencies of ARM play particularly well are Web front-ends and Hadoop environments, we have “ARMed” key partners like Canonical and Cloudera with units.
Early days
At this point it is still early days in the world of ARM servers so we designed Copper specifically for developers and customers to create code and test performance, not for production. To help developers get started we have struck a partnership between the Dell Solutions Centers and Texas Advanced Computing Center (TACC) to provide devs with remotely accessible clusters to develop and collaborate on. And speaking of developers, Dells own devs are working to deliver an ARM-based version of our open source infrastructure management software, Crowbar.
Speeds and Feeds
And in case your wondering about the specs of the hardware:
Dell Copper servers are a shared infrastructure design, which allows easy deployment and reconfiguration of the sleds.
Each ARM server node draws about 15 watts max power, so the total power draw for a full chassis is less than 750 watts.
The server nodes discover themselves and interconnect when deployed, so workloads can easily run across the entire 48 nodes.
And it’s still powerful, with four ARM server nodes per sled, and 12 total sleds, bringing a total of 48 server nodes to a single 3U C5000 chassis.
As I mentioned in my last entry, Mark Shuttleworth of Ubuntu fame stopped by Dell this morning on his way back from CES. Between meetings Mark and I did a couple of quick videos. Here is the second of the two. Whereas the first focused on the client, this one focuses on the Cloud and the back-end.
I’m always interested in what’s happening at Canonical and with Ubuntu. Last week at Hadoop World I ran into a couple of folks from the company (coincidentally both named Mark but neither Mr. Shuttleworth). Mark Mims from the server team was willing to chat so I grabbed some time with him to learn about what he was doing at Hadoop World and what in the heck is this “charming” Juju?
Some of the ground Mark covers
Making the next version of Ubuntu server better for Hadoop and big data
(0:34) What are “charms” and what do they have to do with service orchestration
(2:05) Charm school and learning to write Juju charms
(2:54) Where does “Orchestra” fit in and how can it be used to spin up OpenStack
(3:40) What’s next for Juju
But wait, there’s more!
Stay tuned for more interviews from last week’s Hadoop world. On tap are:
Yesterday at Cloud Expo I bumped in to Dr. Rich Wolski, CTO and co-founder of Cloud player, Eucalyptus. It had been a while since we had last talked so I grabbed some time with him and got him to give me the skinny:
Some of the ground Rich covers:
Eucalyptus’s major release which is coming out in the next 4 weeks
[0:40] The RightScale myCloud integration that they announced yesterday (linking Eucalyptus private clouds with various public clouds)
[2:01] Eucalyptus’s relationship with Canonical and how their interests are diverging
[3:15] Where specifically Eucalyptus is targeted
[4:25] What are some of their goals and product features they’d like to add over the next year
Today when I was walking the floor at the Cloud Expo here in New York, I ran into fellow Austinite Dustin Kirkland. Dustin is the manager for systems integration team for Ubuntu. I got Dustin to give me the low down on the most recent UDS (Ubuntu Developer Summit) that concluded a few weeks ago in Budapest:
Some of the ground that Dustin covers
The big areas of focus on the server side coming out of Budapest
Getting behind OpenStack as the Ubuntu IaaS platform
[1:09] The pioneering work they’ve done with Eucalyptus and how its use case differs from that of OpenStack
[2:05] The Ensemble project, a service orchestration framework for the cloud which is the brainchild of Mark Shuttleworth.
[3:59] Ubuntu Orchestra for cloud installation, provisioning and configuration management (using Puppet)
Yesterday, the announcement went out that the Dell | Canonical Enterprise Cloud, Standard Edition was out and ready for consumption. What this cloud-in-a-box allows folks to do is to set-up affordable Infrastructure-as-a-Service (Iaas)-style private clouds in their computer labs or data centers. The cool thing is that, because the Ubuntu Enterprise Cloud (UEC) software is compatible with Amazon Web Services EC2 and S3 services, it enables IT admins and developers to move workloads between public and private clouds.
Who cares?
Application developers and IT service providers and admins who are setting up cloud POC’s are perfect candidates for this pre-configured testing and development environment. With regards to industries, areas where there is a lot of software development work like Hosters, Telco & Communications, Media & Entertainment and Web 2.0 businesses are prime markets for the Dell UEC solution.
So what’s in it?
The solutions’ basic components are Dell PowerEdge C systems plus a Dell-specific download of the Ubuntu Enterprise Cloud (made up of the Ubuntu operating system and the Eucalyptus platform for private cloud computing). To simplify getting the whole shebang up and running Dell and Canonical are providing the following:
Walrus Controller (W) – the cloud’s storage repository
Cluster Controller (CC) – the controller for a up to 1024 compute cores grouped together as a cluster
Storage Controller (SC) – the controller for cluster’s storage repository
Compute Node (CN) – cloud’s compute node
And on the support side…
If you’re looking for systems management and support services with your order, you are in luck. Canonical, the company behind Ubuntu, has put together UEC Assist, a support service built specifically for Dell customers deploying SE Edition and which is delivered by Canonical’s Global Services and Support team.
Its all about efficiency
From a Dell DCS (the group at Dell behind this) point of view, this offering fits in well with our strategy of bringing total solutions to market that optimize efficiency at every layer, from code to servers to storage. The open source Dell UEC solution is tailor made to deliver a ready to go IaaS solution.
After the cloud summit last week at OSCON, I sat down with Neil Levine of Canonical to see what was in store for Ubuntu cloud-wise (Canonical is a partner of ours in our cloud ISV program). Neil is the VP of Canonical’s corporate services division which handles their cloud and server products.
Here’s what Neil had to say:
Some of the topics Neil tackles:
The next Ubuntu release “Maverick Meerkat” and its geek-a-licious launch date: 10.10.10.
Look for Maverick to make Eucalyptus even easier to deploy and use.
Data processing and data analytics is one of the key use cases in the cloud and Canonical is looking to move up the stack and provide deep integration for other apps like Hadoop and NoSQL.
What are some of the areas of focus for next year’s two releases i.e. 11.04 and 11.10.
Project ensemble: what it is and what its goals are.
Rick Clark used to be theengineering manager at Canonical for Ubuntu server and security as well as lead on their virtualization for their cloud efforts. He’s now at Rackspace and is applying much of what he learned while at Canonical to his new gig as project lead and chief architect of the just announced OpenStack Compute.
Rick talked to me about what he brought with him from Canonical as well as the details behind OpenStack Compute.
Some of the topics Rick tackles:
What is the OpenStack Compute project (hint its a fully open sourced IaaS project)
Leveraging what Rick learned from the Ubuntu community, including a regular six month cadence.
Rick’s goals for design summit: develop a roadmap for the first release, spec out the software and spend the last two days prototyping and hacking.
Why they went with the Apache 2 license and why not AGPL?
The Rackspace API (NASA had already started to switch from the Amazon API before combing
As a follow on to last week’s PowerEdge C line overview, here is the first individual system overview: the C6100. Click below and let Dell Solutions Architect Rafael Zamora guide your thru the design and features of this densely packed machine targeted at HPC and cloud workloads.
Some of the highlights:
The PowerEdgeC 6100 holds the equivalent of 4 systems which have been packaged into “sleds,” each containing boards, RAM and microprocessors.
Upfront you can put a ton o’ disk drives, either 24 x 2.5″ drives or 12 x 3.5″ drives.
Great for markets like HPC clustering and search engines where compute density is key. (This is not intended for running general purpose apps like Exchange, SQL or Oracle).
It will serve as the compute node in the Ubuntu Enterprise Cloud solution from our partner Canonical.
Today is the big day. The one we’ve been working towards for a long time. As a part of Dell’s quarterly launch “Solutions for the Virtual Era,” we are announcing the introduction of:
Dell Cloud Solution for Web Applications, our first Revolutionary cloud solution. Partnering with enterprise-class cloud pioneer Joyent, we are offering a turnkey private PaaS solution comprised of pre-tested, pre-assembled and fully-supported hardware, software and services — all sold and supported by Dell. This integrated solution is targeted at enterprise app developers who are looking to develop “new world” applications in the cloud to be deployed in the cloud.
The Cloud Partner Program Working with cloud ISVs we will be offering easy-to-buy and deploy cloud solutions and blueprints optimized for and validated on Dell platforms. The first three partners we are announcing are Aster Data (providing web analytics), Canonical (offering an open source Infrastructure-as-a-Service private cloud) and Greenplum (self-service data warehousing). (On the Evolutionary cloud side we will continue to work with VMware and Microsoft and stay tuned for news on what’s happening on the Windows Azure front .
A new line of hyperscale-inspired PowerEdge C servers including the PowerEdge C1100, C2100 and C6100 targeting HPC, data analytics, gaming and cloud builders. These are based on the designs we have created for the some of the worlds largest internet companies and cloud providers.
A suite of cloud professional services to help customers assess, deploy, design and manage dedicated solutions.
This is just a quick overview of what we are announcing today. Stay tuned for more details and info in the coming days and weeks.
Extra credit reading: Our Cloud solutions press release
My favorite cosmonaut-coder Mark Shuttleworth stopped by our offices this morning for a visit. Mark is the founder of both the Linux distribution Ubuntu and its commercial sponsor Canonical. Mark and I sat down in the lobby and caught up. Here is a short interview we recorded.
The 10.4 Ubuntu release Lucid Lynx and what to expect: a strong cloud focus on the enterprise side and a lot of shiny new bling on the desktop as well as making the desktop “social” (e.g. Tweet straight from your desktop)
What Ubuntu is doing in the Netbook space
What excites Mark the most in technology today and why cloud is like HTTP in the early 90′s
Mark Shuttleworth, founder of Ubuntu and the head of Canonical, the commercial entity behind the popular linux distribution, is currently making his rounds in the States. Yesterday he was quite busy, taking the stage at both the Intel Developer Forum in San Francisco as well as at LinuxCon up in Portland Oregon.
Today he popped by Dell here in Austin to chat. I grabbed him for a few minutes right before lunch. Here is the result:
Yesterday I attended a webinar that RightScale put on entitled: How to Build Scalable Websites in the Cloud. It was basically a welcome to RightScale, welcome to the cloud presentation but overall interesting and credible.
The presenters were their CEO, their head of marketing and a mini team of techies. Below is part one of some of my thoughts and takeaways. But first a slight digression…
Enter the Dolphin Master
One thing I noticed during the presentation and which warmed my heart was that MySQL played prominently in a bunch of the slides. It was only today when I was poking around the RightScale site that I saw the press release from a few weeks ago announcing that Marten Mickos, former MySQL CEO and Sun employee joined the RightScale board of directors. Its interesting but not surprising to note in the release that Marten calls out Sun and Canonical (the commercial sponsor behind Ubuntu) as two strategic partners helping to expand the RightScale ecosystem.
Where Right Scale fits within the tri-sected cloud.
Where they play in the Cloud(s)
RightScale positions themselves as a cloud management platform or as I like to think of it “a cloud tamer.” If you split the cloud in three — software as a service, platform as a service and infrastructure as a service — they play in the last space. Basically Right Scale sits on top of Infrastructure as a Service (IAAS) and can handle all the tricky bits so you don’t have to.
Choose or choose not to choose
For those who want more control over their infrastructure RightScale will allow you to “choose among a variety of development languages, software stacks, data stores and cloud providers.” For those less intrepid in the cloud they have server templates that you can start off with.
One of the key benefits they stressed was getting rid of vendor lock-in, “so that you never get locked in to a single provider.” You’ll notice on the X axis above they show lock-in decreasing and portability increasing as you move to the right. My question however is that with Right Scale aren’t you simply locked in to a different layer of the cloud? Doesn’t the control point simply move up the stack? Just wondering…
At Austin Cloud Camp on Saturday I ran into Ubuntu linux developer and Canonical employee, Dustin Kirkland. Dustin is on the server developer team at Canonical and, as he explains it, focuses on various aspects of virtualization, the plumbing layer below cloud computing. I grabbed Dustin for a few minutes and chatted with him about last week’s release and what he’s been working on.
Michael Cote of Redmonk welcomes us all. (credit Dave Nielsen)
This past Saturday, Cloud Camp Austin was held down on the UT campus. There was a very healthy turnout and a lot of great discussions were generated.
Sequence of Events
After opening salutations, camp got underway with a series of six five-minute lightening talks delivered by the camp’s gold sponsors.
My lightening talk: Mapping Processes in the Cloud (credit: Dave Nielsen)
The Main Event
From there, Dave Nielsen, the man who originally developed the cloud camp format (and who took most of the pictures in this post — see them all), guided us through the process of coming up with topics for session discussions. That process, appropriately enough given that this was an unconference, began with an “unpanel.”
The Unpannel: Michael Wilde of Splunk, (not sure), Dustin from Canonical, Cote, myself (credit Dave Nielsen)
All Together Now
The way it worked was the room first brainstormed a list of topics they were interested in discussing/learning more about. Anyone who thought they were an expert on one or more of these topics got to get up from their seats and form a five person panel at the front of the room. Each member of the panel then answered two questions from the board and as the question was answered the audience was asked if the topic had been covered by the answer or if it warranted further discussion in an afternoon session
(L->R) Dave Nielsen leads us through our unconference set up. Canonical's Dustin Kirkland and hero-for-hire John West lend a hand.
A Schedule is Born
After the panel, as a group we all decided what the final sessions would be and who would lead them. To lead a session you could either be knowledgeable in the area or completely clueless but wanted to learn about it.
The completed schedule: three sessions ran at a time and there were three time slots (credit Dave Nielsen)
Coming Soon
While at camp I did a couple of video interviews, one with Dustin Kirkland of Canonical and one with Todd Morey of Mosso/Rackspace. I should be posting those in the next few days. I also found myself on the other end of the microphone being interviewed by Mr. Cote. That should be appearing in the near distant future on his blog.
Posted by Barton George 
.
