Archive
OpenStack’s First Year: How a Committed Community Made a Difference
.
You know the saying: “Time flies when you’re having fun.”
This week has been a crazy one, so I didn’t get to chime in on Tuesday with my thoughts on the one year anniversary of OpenStack.
So today, I took some time to think back over the last year, and I realized how far we’ve come as a technology and as a community.
In addition to the solid OpenStack technology that is being guided by market requirements and pushing the envelope, I believe that the unique set of OpenStack developers, partners, and users has helped get it to where it is.
Pardon me, while I stroll through memory lane…
Design Summits
- Austin Design Summit (July 13 – 14, 2010) – I’m not sure if we could have called this one a design summit as it was more of a meeting of minds, lots of ideas, and a ton of excitement about this new platform called OpenStack. Got to hear from both Rackspace and NASA on the code bases, and how this could change the market. I specifically recall our group of cloud solution attendees from Dell (the company I work for) talking about how much potential this technology had. (And besides, this meeting was in our home town!)
OpenStack was announced publicly for the first time a few days later on the 19th. Dell was among a handful of companies who believed in the initiative back then – it was early – and we had to have vision.
- San Antonio Design Summit (Nov 9 – 12, 2010) – This was held at the Weston Centre in San Antonio, and I remember thinking “where did all these people come from???” We had a lot of international presence there from the UK, France, Japan, and other parts of the world. It was exciting to think that in just four months, we’d already started going global. I also recall we started talking to the business of OpenStack – licenses, brands, etc, which was a good sign of progress.
Dell did our part as well – Rob Hirschfeld and Greg Althaus, a couple of Dell OpenStack rockstars, each spoke on OpenStack deployment, reference architectures, and operational models. We also contributed gear for the InstallFest later that week, made up of PowerEdge C6100s and C2100s (I also remember giving quite a few tours of the “server room” to see what Austin release was running on.)
- Santa Clara Design Summit (Apr 26 – 29, 2011)– Our first design summit on the west coast! This one is still pretty fresh in my mind, but what blew me away was again, the growth! The community grew and became more sophisticated. Those that were learning at the first two design summits were blossoming experts at this design summit, having lived through the evolution of OpenStack. Though I’m sure it was logistically nuts for the organizers as they greatly exceeded capacity, it was a great indicator that we were on to something special.
The Dell team gave it our all here as well – Rob gave a session on what Dell was doing with OpenStack and Crowbar, and my favorite – daily live demos of Crowbar deploying multi-node OpenStack clouds on bare metal servers! (I’d time our guys – the best time was 29 min minutes for a 6 node OpenStack cloud running on Dell PowerEdge C 6100’s – not too shabby!
(Anyone remember the Crowbar bunny shirts?)
Year One Partners
I just checked the OpenStack.org site, and at THIS MOMENT, there are 91 partners in the community. (Depending on when you read this, there could / will be more.)
You’ll find a number of key industry players there – Rackspace, Dell, Citrix, Intel, AMD, Cisco, Canonical, Brocade, Arista, Opscode, and more. And this group has done a lot to further the intiative over the last year.
Here are a list of a few examples.
- Rackspace announcing the creation of Rackspace Cloud Builders, who’s purpose in life is to service customers on OpenStack deployment, training, support, and consulting.
- Citrix announcing Project Olympus and a distribution of OpenStack to come soon
- Dell announces the Crowbar deployment software for OpenStack, and our intention to release an OpenStack solution to market
- Canonical announces their intention to make OpenStack the default cloud platform in the Ubuntu operating system
- Equinix’s sponsoring of a live OpenStack demo enviornment w support from Dell, Rackspace, and Citrix
- Real live production usage by companies like Internap
What a year.
I’d venture to say those of us who were there on Day 1 believed this was going to be big, but we’re excited that it has been adopted by the cloud community as much as it has.
And to all the partners, develeopers, and users who have made the first year amazing – I salute you.
We’re on to something big. 🙂
If you want to learn more about what Dell has done with OpenStack over the last year, and see if what we’re building is a fit for you, email us at OpenStack@Dell.com.
More info on OpenStack and the one year anniversary:
- The GREAT infographic from Rackspace on the past year! http://www.rackspace.com/cloud/blog/2011/07/19/happy-1st-birthday-openstack/
- Rackspace’s Lew Moorman: http://www.rackspace.com/cloud/blog/2011/07/19/what-a-year%C2%A0-openstack-at-one/
- Jim Curry’s blog on OpenStack.com http://www.openstack.org/blog/author/jimcurry/
- Derrick Harris’s great article on what’s next http://gigaom.com/cloud/openstack-turns-1-whats-next/
- The OpenStack.org Design Summit page – lots of historical info and pics http://www.openstack.org/blog/tag/design-summit/
- Dell’s Rob Hirschfeld’s (@zehicle) tech blog – a ton on OpenStack – http://www.RobHirschfeld.com
- The Dell OpenStack site featuring the popular technical whitepaper “Bootstrapping OpenStack Clouds” www.Dell.com/OpenStack
Until next time,
JOSEPH
@jbgeorge
Cybera, OpenStack, Rackspace Cloud Builders and Dell
.
Recently, Cybera, a non-profit Canadian outfit who’s core objective is to drive innovation among Canada’s tech community, recently blogged about their path to production, open source cloud instance based on OpenStack, with their first installment being on the topic of hardware.
Dell and Rackspace were fortunate to be called out as innovators in helping them on the path to the OpenStack cloud.
Here’s a quick excerpt on their server hardware choices:
We ordered four different types of servers (aka nodes). A management node (nova-api, nova-network, nova-scheduler, nova-objectstore), compute nodes (nova-compute, nova-volume), a proxy node (swift-proxy-server) and storage nodes (swift-object-*, swift-container-*, swift-account-*). All nodes were contained in the Dell C6100 chassis. Here are the specs:
Processor Sockets Cores Threads RAM Disk Management E5620 2 8 16 24 8 x 300 GB Compute X5650 2 12 24 96 6 x 500 GB Proxy E5620 2 8 16 24 4 x 300 GB Storage E5620 2 8 16 24 6 x 2 TB The disk on the compute nodes is used for VMs and volumes, which is to say:
- a portion can be used for VM instances, the files that back the VMs
- a portion can be used for volumes, the files that back the virtual hard disks for the VMs (technically speaking it’s logical volumes that back the virtual hard disks but you can think of them as files). See Managing Volumes.
.
It’s a great read, and I highly recommend anyone even remotely considering OpenStack to take a look, and follow them as they continue to provide the community updates on their journey. (The link to Cybera’s post is available at the end of this blog entry.)
As Dell announced a few months ago, we’re actively engaging telcos, hosters, service providers and others on getting them going on OpenStack today. We have OpenStack experts on staff developing product, working with the community, and driving innovative design – you probably recognize some of the names – Rob Hirschfeld, Greg Althaus, and a few others.
And we’re proud to say that Dell was the ONLY hardware solutions provider to have been part of the community since its creation in July 2010.
To date, we’ve
- Authored the wildly popular “Bootstrapping OpenStack Clouds” technical whitepaper, which details hyperscale design for multi-node OpenStack clouds
- Developed the as popular OpenStack tool, Crowbar, which deploys a MULTI-NODE OpenStack cloud in a matter of minutes rather than hours (yes, that “MULTI-NODE” differentiation is important). It also allows for certain infrastructure configuration, network discovery, and is the basis for the operational model we feel will be optimal for OpenStack management.
- Defined a reference architecture for OpenStack on Dell PowerEdge C cloud optimized servers
- Been developing relationships with a number of valued partners in the OpenStack community, such as Rackspace Cloud Builders, Citrix, Opscode and others, as we continue to get customers to a full OpenStack cloud.
If you’re looking to start on the path that Cybera has gone down with OpenStack as a cloud platform, we’d love to talk to you. Reach out to us at OpenStack@Dell.com.
And stay tuned – the best is yet to come. 🙂
More info:
- Cybera’s blog – http://www.cybera.ca/tech-radar/running-openstack-production-part-1-hardware
- Dell’s OpenStack website (including link to “Bootstrapping OpenStack Clouds” whitepaper) – http://www.Dell.com/OpenStack
- Video on Dell’s Crowbar tool – http://www.youtube.com/watch?v=4zgT-6G2fXA
- Rob Hirschfeld’s blog – http://www.robhirschfeld.com/
Until next time,
JBGeorge
@jbgeorge
Cloud Driving Change
Those of you that follow me here at the JBGeorge Tech blog or on Twitter or in other ways (stop stalking me!) know that one of my passions is how technology can make a difference in the world. It can be through technology leaders advocating philanthropy, tech innovation leading to medical advances, or a local IT person donating their tech talent to improve school safety.
To that end, I was thrilled to be invited to a meeting that took place in San Francisco yesterday.
On Monday, I, along with a few other select individuals involved in the OpenStack open source cloud initiative, had the distinct privelege of meeting with Aneesh Chopra, CTO of the White House, to discuss cloud computing, open source, and how open source cloud (specifically OpenStack) can help America grow and thrive.
A number of key representatives from various OpenStack participants were present from Dell, Rackspace NASA, Citrix, AMD, and others to contribute to the discussion. There was a lot of great conversation, which included an overview by the OpenStack team from Rackspace, and a myriad of perspectives from the various attendees. Specific problems were discussed such as security, standards, and impact on other areas like healthcare. Much was discussed about OpenStack in particular, and how its evolution can help spur on the American economy.
What are your thoughts on how cloud can better society? How does it impact education, healthcare, or small business? How can it make us healthier, economically more stable, and overall better citizens? Let’s think beyond operating systems and PaaS / SaaS – what can cloud do to help drive positive change?
Regardless of where you fall politcally, it’s a great time to be in technology. The current administration views tech and cloud computing as a key driver to advancing America and rightly so.
And they’re looking to us to help figure out ways to do that.
That’s a real cool thing. Let’s pitch in.
I’d be interested in hearing your thoughts. Feel free to drop me a comment or reply on Twitter @jbgeorge.
Until next time,
JBGeorge
www.jbgeorge.net
@jbgeorge
Thoughts from 2010 Gartner Data Center Conference (Part 2)
Hello all – hope you’re having a good Saturday / Sunday wherever you might be.
Wanted to finish putting down thoughts, insights, etc from my time at the Gartner Data Center conference this past week. (You can read Part 1 here – https://jbgeorge.net/2010/12/11/thoughts-from-2010-gartner-data-center-conference-part-1/.)
- We need to understand the success / real world utilization of ITIL and other benchmark frameworks – are they working?
- More and more, in the era of cloud, we are finding it is no longer necessary to keep an individual system up at all costs, as long as overall compute and storage integrity are maintained
- Traditional management models assume that systems should be managed so that failure should rarely happen. Newer models assume that failure WILL happen, and focus on shortest MTTR (mean time to recovery / repair).
- Traditional models try to implement pervasive automation, whereas newer models focus on selective automation. Why must we automate / virtualize / etc everything? Choose wisely based on criticality and true need.
- We’ve heard of JEOS – the “just enough” operating system. Gartner spoke of “just enough” practice vs “best” practice. Are we at the era of “just enough?”
- Again, reiteration of the need of DevOps skillset.
- Organizational alignment is still a key facet of moving the IT organization.
- “We are only at the end of the beginning” of the cloud era. Watch for Cloud 2.0 in the years ahead (market based computing, hybrid clouds the norm, etc)
- Still a lot of talk about the Big Four (HP, CA, IBM, BMC) – they were slow to jump on w virtualization, but more aggressive with cloud.
- Definite focus on the network being a key management focal point. Similar to the theory that your band’s ripping concert is only as good as the quality of your sound man.
- The recession will be viewed in hindsight as a pivot event for the server market – paradigm shifts, vendor repositioning, etc.
- Some important trends to watch going forward: big data, unified communication, client virtualization, compute density / scaling vertically, converged fabrics
Another great event – look forward to next year.
Until next time,
JBGeorge
@jbgeorge
Thoughts from 2010 Gartner Data Center Conference (Part 1)
This week, I had the pleasure of attending the 2010 Gartner Data Center conference – got to see a lot of old friends, meet new friends, and learn a lot about what Gartner sees coming down the road.

This year's Gartner Data Center Conference was held at Caesar's Palace, Las Vegas, NV
It was also a chance to talk to a number of folks about what’s happening at their own data centers, what they’re looking to solve, and what they’d like to see start happening in the industry.
Here are some key nuggets I walked away with – I’ll post again tomorrow with the rest:
- When it comes to implementing cloud, we cannot allow “20th century industrial models to sap 21st century innovation.”
- There’s still not a good answer for failure remediation in the cloud – credit due to downtime is just not good enough.
- Expect the community cloud concept to continue to draw interest. (Community clouds are clouds that service specific areas like banking or healthcare, where compliance, etc would be a requirement for its customers.)
- The next big business opportunity could be cloud brokers as the new systems integrators
- Great quote from Phil Dawson regarding due diligence before virtualizating anything – “Don’t virtualize rubbish – otherwise you have virtual rubbish.”
- We often forget that virtualization is more than just servers and storage – there are apps, desktops, etc
- Client virtualization / VDI is still top of a number of minds, though many are still at the investigative stage. There are still lingering questions about user adoption, bandwidth / network constraints, and ROI. (Though I am a big believer.)
- When we build staffs, we should strive for them to be “T-shaped” – technically deep in few areas, but linkages to the broader business.
- It’s important to run IT as a business – remember that it is providing something of value that its customer is willing to pay for
- Some good discussion on IT chargeback and allocation, which many are not doing today, but forsee implementing in the future. Four required characteristics of IT chargeback: simplicity, fairness, predictability, and control.
Also got a walkthrough of the IBM containerized data center, as well as SGI’s container – both very cool. (No pun intended.) I’ve now had the pleasure to see the modular / container data centers from HP, Dell, IBM, and SGI first hand.
Some interesting stats and statistical predictions from Gartner:
- 2/3 of the live audience was polled said they will be pursuing a private cloud strategy by 2014
- What’s the top concern regarding cloud computing? Security and privacy are still at the top.
- If Facebook were a country, it would be the third largest in the world. Twitter – the 7th. (Wow.)
- There has been more video uploaded to YouTube in the last 2 months than if ABC, NBC, and CBS had been airing content 24/7/365 continuously since 1948. (WOW.)
- Data centers can consume 40x – 100x more energy than the offices they support.
- An 8,000 square foot datacenter could cost $1.6M per year for just power.
- Data centers will be significantly smaller in the next 5 – 10 years
- Data expected to grow 800% over the next 5 years, and 80% of it will likely be unstructured.
- Today’s labor force will have 10 – 14 jobs by age 38
As you can tell, just a lot of good discussion on cloud, data centers, power, and overall IT.
OK, don’t want to overload more than I have – will back tomorrow night.
(UPDATE: Click here for Part 2.)
Until next time.
JBGeorge
@jbgeorge