 We got the mics on. All right. Good morning. We're going to get started here shortly. Thanks for joining us this morning. I'm Jesse Proudman, CTO of Blue Box. And I am Shamal Tahir, IBM Private Cloud Offering Manager. And we're here to talk about OpenStack for beginners and to kick off the IBM Cloud Client Day. Thank you all for coming. So Let's start with the history of OpenStack and where it originated. So a long time ago in a city far, far away, Austin to be precise, OpenStack was created and launched by a joint project through NASA and Rackspace. So in 2010 the project was created and in the beginning OpenStack had object storage and compute services. But then as the needs for those services from a, you know, elastic provider perspective grew, more and more companies and contributors started joining and contributing code and eventually we realized that, you know, SUSE, for example, announced the first public preview in 2011. By 2012 there were more and more contributors to OpenStack coming up and so the need for a foundation was created. And so in 2012 we took the project that was started by NASA and Rackspace in 2010 and we made it into a open foundation, the OpenStack foundation as we know it today. Also, in 2012 Red Hat introduced their OpenStack distribution. In 2013 Oracle became a sponsor and 2014 Helion was announced. So as you can see the trend is that we've had growing number of community and as of 2015 we have over 3,900 developers and about 31,000 members in the OpenStack foundation community. So it's interesting to me about this history is that you can see from very early on you took an open source project that had a pretty, pretty good community contributing and very quickly grew commercial support, not just from one company, like we often see in the marketplace today with other, other open source projects, but from a whole diverse set of companies. So the need for OpenStack, the need for this type of technology was evident from very early in its evolution. Absolutely and I should mention that in 2012 when the OpenStack foundation was created IBM was a platinum member from the beginning of the Foundation as well. Absolutely. So IBM was in since 2012. So I'd love Jesse to explain you know when BlueBox joined OpenStack and what the drivers were there. So BlueBox was founded in 2003 and the company grew up as a managed hosting company. So we were helping our customers deliver their applications to their clients. In 2010 we began to see a shift in the marketplace and a shift in customer demand. Customers were beginning to ask for self-service as the result of what was happening in the public cloud space. They were asking for APIs. They wanted to automate their infrastructure. At the same time we saw those customers looking for a private cloud solution, something that wasn't shared amongst the many other customers. And so in 2013 we launched what we call BlueBox Cloud at the Hong Kong OpenStack Summit. BlueBox Cloud at that time was a hosted private cloud. So it was a dedicated offering to a single customer operated out of our data centers. We had four globally delivered to the customer where BlueBox managed all of the OpenStack all the OpenStack operations for the customer and the customer just worked with OpenStack. At that time we were on the grizzly lease and the offering has been evolving ever since then. Thank you. So let's talk a little bit about what were the business drivers that customers were seeing that made them think that you know cloud is the next platform for our data center and our evolution of how we leverage technology for our business. So the data points here by the way are from the OpenStack Community User Survey. And so every release OpenStack does a survey where they ask the community operators of OpenStack and consumers of OpenStack services what do you use an OpenStack and why do you use it. And so from that perspective one of the notions that was early on within the OpenStack and OpenSource Cloud movement was it's a cost saving initiative. And it is cost related but really the true power of cloud now that we're more mature in how we're enabling it integrating into our business has become the increased operational efficiency and the agility that it provides. So saving money is good agility and being able to build better products, more scalable products and faster time to market is the real driver behind cloud. Well that's interesting when you think about saving money what does that actually mean. I think many people originally looked at OpenStack as a way to save money on VMware licenses or other virtualization platforms. And sure if you can tolerate sort of the cloud native aspect of OpenStack that's a viable replacement. But at the same time there's an opportunity to use OpenStack as a platform to build your next generation applications, your cloud native applications directly. And it is through that process that customers are really saving money. Being able to build, deploy and operate an application in a much faster mechanism because of the power of OpenStack and its APIs because of the cloud capability that it delivers. That saves customers money. And I think it saves customers more money than just trying to replace existing infrastructure. We were on a panel yesterday where we were talking about is OpenStack a replacement for VMware? Does it collide with what VMware is doing? And I think the answer is no. I'm a big believer in it. If it's not broke, don't fix it. And so customers don't want to spend the energy to move something off of VMware that's already working. But they do want a platform that can sit adjacent to that VMware infrastructure that can sit adjacent to their existing data. And they can build net new applications on top of. And that's what OpenStack can provide. So now that we've talked about some of the business drivers, let's talk about some of the actual business applications that people are building on top of OpenStack. And so throughout the years we've had, as you've seen actually in the OpenStack conferences, every year more and more users are actually part of the keynote. In the beginning it used to be the foundation and vendors and now it's actually the users that drive most of the content during keynotes. And a lot of that is because now OpenStack is being applied technology to actual business problems. And so from that perspective e-commerce is a very well-known use case. Subu Almanjaro from PayPal is the user committee lead or chair now. And so eBay is using OpenStack daily. There's other retailers worldwide that use OpenStack for hosting sites that literally get millions of hits per day. So companies like Walmart from the United States, Target, these companies have decided to base their e-commerce platforms on top of OpenStack and make a tremendous difference to the velocity of their business. Absolutely. And another interesting one that's new is this notion of agile platforms for applications. So in the last keynote in Vancouver we had a company called Digital Flimtree come on stage and talk about how they're doing a digital supply chain for video editing, production and delivery using OpenStack. And actually one of the big enablers around that is the new federate identity feature within OpenStack itself. Being able to have intraoperable clouds that have common API and common response to be able to work on applications and data sets in one cloud and be able to use those same applications and data sets in other clouds as you move from vendor to vendor in your supply chain workflow. Digital Flimtree I think is a fascinating customer in that they're using OpenStack in a way you may not think about. So I had the opportunity to go down a few weeks ago and meet with the CEO of Digital Flimtree and hear more about their experiences. Digital Flimtree edits many of the popular TV shows in the US and they collect a lot of video footage for each show that they're working on. They're able to take that video footage and it's all a file based workflow now, so no longer on tape, it's all digital files and they're using Swift from OpenStack to store that data. But they're not using Swift from a public cloud provider. They've got Swift installations that actually sit in their office where the production work is done. They've also taken Swift and put it on an appliance. So if they're shooting a TV show in the field, they could take that appliance out in the field and actually put it next to where they're shooting. And so OpenStack doesn't just mean a private cloud in somebody's data center. The technology is flexible enough that you can use individual components to change the way your business operates. And I think that's a pretty exciting component of what's going on in the OpenStack space today. Absolutely. And this is the part I love about OpenStack is, you know, it's just not doing the same existing business applications and processes. It's adding new capabilities and new processes that we might not be able to even imagine in the past. Opening up new doors that weren't available before. Exactly. And the other point I want to make about this one, as I mentioned, Federated Identity was one of the enablers for this type of a unique use case. And so BlueBox was one of the first 15 companies to commit to supporting Federated Identity to have interoperable OpenStack clouds. And so over the next year, I think you'll begin to see those product offerings are allowed from those 15 different companies. That release will begin to allow companies to use multiple vendors, all powered by OpenStack, to deliver their applications and to share a similar set or same set of credentials across all of those offerings, which makes OpenStack and all of its global points of presence, whether it's from BlueBox and IBM or from other vendors, a lot more flexible than some of the other offerings in the market. So moving on to the next use case. This is another popular one, which is HPC. And so, again, in Vancouver, we had another user, CERN, come on stage and talk about their environment. And CERN actually operates the LHC, the Hydron Collider. And they actually do research and data analytics on those collisions on some OpenStack powered clouds. And also CERN is known for their scale. Their cloud is upwards of 50,000 cores plus. And so HPC has been a long-standing use case. And we all know how intensive HPC can be. Well, and it's fascinating to look at the projections that CERN is considering as they think about all the data that they collect and analyze. So over the next ten years, the scale that they'll require far exceeds that 50,000 core count into the hundreds of thousands of cores and petabytes of data. And having selected OpenStack as their platform, they are highly motivated to work with the community to figure out how to continue OpenStack's capability to support that type of scale and that workload. And then the last workload that we want to highlight is, you know, the cloud-native workload that Jesse mentioned earlier as well. So enabling web services, enabling web applications at scale, distributed, oriented in a microservice architecture, OpenStack was built with that type of application architecture in mind. So it's naturally suited to basically build the next unicorn. You know, we see this type of deployment with a lot of our customers and a lot of our prospects. Particularly with the Blue Box Local product that we announced yesterday where we can put a private cloud in a customer's data center. This type of deployment allows customers to build their next-generation applications, but have those applications talk to their existing system of the record, their existing databases, their existing file stores. And so you're able to leverage the investments you've made today, but build your next-generation applications in a modern architecture to be able to future-proof your investments. We see this in banking, we see this in healthcare, we see this in finance. Those types of organizations may not be able to move their mainframe systems, may not be able to move their transaction engines, may not be able to move their customer data stores from where they are today on premises into the cloud, but they want to be able to leverage OpenStack. They want to be able to build new generation applications using cloud. And this type of approach solves that problem. Absolutely. And so we've explored maybe four or five different use cases that OpenStack has been applied to business problems. In reality, because of its flexibility and versatility, it can be applied to probably a hundred different more use cases that we don't have time to cover right now. The other thing to add is because it is SilverStole, it's a foundational level service, so as you develop applications like web services, you know, you get into BlueMix, and even BlueMix can benefit from OpenStack underneath it, so. Absolutely. So now that we've talked about the value, and you know, we're all excited that this actually makes sense from a business perspective, so getting started should be simple, right? Let's go make this happen and harness the power of OpenStack. Well, that's been the challenge historically, is it's not simple to start. And the challenge has been, in the past, OpenStack used to have what they used to call the OpenStack integrated release. And that meant that you had projects that were tested with each other to make sure that they didn't break each other's code as they were enhanced release to release. But because the term was integrated, people often thought that I had to download and use all of those projects together. And trying to install all those services when not all those services were needed, made OpenStack really complex from an installation perspective. And this is where, you know, basically a lot of distributions popped up saying, I can help you with the deployment challenge. But as we know, and as Jesse will expand on, deployment was not the right problem to solve, it was operations all along. So if you look at the historical approach to using OpenStack, there have been a couple of methods. The first was to go take the open source packages, the open source available in the repositories, download that and install it yourself. And you see a lot of organizations started out that way, and even some of the largest organizations today still use that approach. The challenge with that approach is that the packages and the software that is in the repositories aren't necessarily architected or configured for the way you need to operate your clouds at scale. Doing highly available deployments, those take sort of a preferential approach that doesn't necessarily come with the source code. So the next thing that occurred was the advent of the OpenStack distribution, taking OpenStack and selling it a software. So in that model, a distributor would take the open source packages, they would put their preferences and configurations into them and ship out those packages to end users. The challenge with that is that it may make installation easier, you may be able to get up and running faster, but the ongoing operations of that environment ultimately still fell on the customer, on the end user. So this is the classic at 3 a.m. when the cloud falls over, who do I call problem? And with just a Linux distribution or a single piece of software, it was less impactful when that system failed. If you had one server down, that may only impact a portion of your infrastructure. But as we think about cloud, most of your infrastructure now runs on that environment. And so when it fails, it can take down your entire business. And so we like to call this the blast radius. OpenStack and a private cloud has a much larger blast radius than a single piece of software. And so that software distribution model didn't solve that operational challenge. And so two or two and a half years ago, you started to see a new model emerge in the marketplace. Bluebox was doing this, MetaCloud was doing this, and more recently, you see companies like Platform9 offering a similar type of implementation, where the vendor does the installation and operations of the environment, and the customer simply gets to work with OpenStack using OpenStack, benefit from OpenStack, without having to worry about OpenStack itself. We call this private cloud as a service or OpenStack as a service. And we think that's the best way for most organizations to consume OpenStack to get the benefits of OpenStack. Couldn't agree more. So let's talk about three takeaways from what we've talked about so far. So the first thing is, as we've talked about already, we talked about the history originally about OpenStack and we said it had compute service and object store. At this point OpenStack, and I'm not even going to go full exhaustive list, has compute, it has object, it has block, it has network, it has message queue, it has DNS as a service, it has workflow as a service. There are so many different components, which is great because as we were talking about, this is what enables the diversity of applications and use cases that OpenStack can support. But flexibility and having this much versatility also leads to complexity. So we have to balance the benefits with the complexity and this is where the private cloud as a service model really shines. I think it's also important to remember that just because OpenStack has all of these projects doesn't mean you actually have to use all of these projects. So if you think about the way OpenStack is organized, there are a set of primitive objects, things like networking, storage and compute. Those will likely be part of your deployment. Those are the essential part of your deployment. But you don't have to when deploying those, take all these additional services. You can pick and you can choose and you can figure out which of those services actually brings maximum value to your business. So the next thing is we have an expression that says you don't want to see how the sausage is made. And so this applies to cloud as well. You can run a cloud, you can operate a cloud, but to be honest, the skill sets that you use to operate a cloud from a distributed systems perspective, from a Python language perspective, those skill sets could be used to benefit your business and generate revenue for you versus support your operations for you. And so this is why you don't want to see how the sausage is made, but you want to be able to enjoy the variety of sausage that is available. So we like to think that it's important for an organization to look at where they believe they can get the most value from the effort that they're putting into their cloud deployments. If you can get the most value from deploying and managing your OpenStack footprint, and some organizations can, many of them are technology companies, then that's a path that's worth exploring. But if you think you can take that same team and have them focus on building your applications, your applications that actually differentiate your business in the marketplace to make your business money, we believe that's a better place for organizations to spend their time. That's the whole premise of public cloud, that somebody else is worried about the infrastructure, that somebody else has figured out all the technology challenges. And so we believe that we can make private clouds have a similar experience to public clouds, but sit in your data center or be dedicated to you in a soft layer facility. Absolutely. And last but not least, as you've seen in the key notes that have occurred and the Liberty release material itself, Liberty is the first big tent release of OpenStack. And what BigTent means is that OpenStack, as long as you are a project team that will adhere to the four opens of the OpenStack community, which is open design, open source code, open discussion, and generally having open communication lines, we will welcome you into the OpenStack family of services. And so with that, that's also going to increase the flexibility and capabilities of OpenStack, but again, it is going to be exponential in terms of the impact it has on the variety of options you need to choose from. And so you can see that even at this summit. So Akonda open sourced their project Astario into the BigTent at this release, which means that we're going to begin to see many more projects entering the OpenStack project space. And so instead of just having an understanding and a comprehension of say 30 projects, that might be 60 or 90 in three to four years. And that adds a lot of complexity to the story. It also adds a lot of flexibility. So it's both a good and a bad thing. And we're looking forward to seeing how that plays out in the OpenStack community. Absolutely. And the project navigator that they announced this week, that is going to be a really great tool as you're learning about OpenStack and the capabilities it offers, I would encourage you to check the project navigator because it has a representation of what services are out there, how much adoption do they have in production, as well as how many different distributions support that service in their OpenStack packaging. So it's just a good amount of data for the maturity of a service. And so that project navigator is available on the OpenStack Foundation website at OpenStack.org. It's in the navigation up top. I encourage you to go check it out. It's a great way to learn more about all of these individual pieces. Absolutely. And then the other piece I would like to add is, as Jesse and Angel highlighted in their keynote, it's Hello World. You know, it's, you can't out-innovate the world. And so what makes OpenStack great is everyone participating in the community. And so what I would like to highlight here is generally, OpenStack does need developers and contributors. As I mentioned, we have about 3900 going on 4000, which will be another milestone for the OpenStack community. But there's business users, there's people that write applications on top of OpenStack Power Clouds, and we want all of those people to participate in the community so we can understand how OpenStack is being used to make your business better, as well as how can we make sure that the user experience in OpenStack continues to evolve. And so from that perspective, there's a ton of different ways you can get involved in the community. Whether it's joining discussions in mailing lists, IRC, whether you share, if you use OpenStack, there's a super user publication where you can share how you're benefiting from OpenStack. We encourage those shares because again, as new people look at OpenStack, those stories help them understand how can I benefit from OpenStack. And so I will be here a little bit after this session ends and if there's ways that you want to understand how you can get involved, I'll be glad to answer those questions as well. Some of the other notable aspects of what the foundation have been working on are things like the user survey, which comes out every six months. And so that's an opportunity for those who are working with OpenStack to provide feedback on what they're using, what they like, what's working and what's not working, which is pretty critical for the foundation and the developers to be able to know where to prioritize efforts. If you think about a regular product or service offering that's provided by one single commercial entity, that entity has a product management capability that's talking to customers, that's talking to the users of that offering. Because OpenStack has so many commercial offerings as an open source project, it takes a lot more effort for the foundation and for the developers to actually know what's going on. And so there's a product working group that's focused on those user requirements. And as Mark noted in the keynote yesterday, the way OpenStack is being architected, the projects that are being worked on in the future, and the bugs that are being fixed are now beginning to be directly derived from the users of OpenStack, which will make the offering much better for those end users. Absolutely. So thank you so much and with that, Jesse, I don't know if you have any closing thoughts, but I think we're done with the content. Really appreciate you all joining us this morning. We've got a whole schedule of IBM Client Day talks here in this room, so we look forward to having you join us throughout the day to learn more about what IBM is doing with OpenStack and the offerings that we have. And look forward to sharing the rest of the week with you. Thank you. Thank you.