We discuss why Vanguard went to the public cloud, the value of DevOps and best practices for IT leaders who are just getting started on their cloud initiative.
Flere episoder fra "Enterprise Initiatives"
Securing Your Containers in the Cloud
21:57Our guest on the podcast this week is Ben Bernstein, CEO and Co-founder of Twistlock. We discuss the myths surrounding the security of applications in containers in the cloud and advice for people just starting a container initiative.
Adopting Cloud and DevOps Across 2,000+ Developers at Vanguard
23:00We discuss why Vanguard went to the public cloud, the value of DevOps and best practices for IT leaders who are just getting started on their cloud initiative.
Multi-Cloud vs Hybrid Cloud: What’s the Difference?
34:05Our guest on the podcast this week is Issy Ben-Shaul, Co-Founder & CEO at Velostrata. We discuss the difference between hybrid cloud and multi-cloud. Hybrid is anything that’s not just one cloud provider, which includes multi-cloud. Multi-cloud is a strategy where some workloads are running on one cloud, some on another. One trend we see is that big enterprises are splitting workloads more and more between at least two major public clouds for a multi-cloud strategy. According to Gartner, 70% of enterprises will be implementing a multi-cloud strategy by 2019. Some do it because they are mandated by regulators to have more than one cloud. Others are deciding they can’t put all eggs in one basket and want the flexibility to use multiple different vendors. Companies used to look at cloud as a data center and it was all about IaaS. Now it’s becoming more and more abstract: PaaS, containers as a service, microservices, functions as a service, IoT developer kits, and edge computing are just the beginning. It seems that the more mature a company is in the cloud, the further up that stack they go. Where is all this going? Customers are looking for operational efficiencies, cost savings and more productivity and agility. In conclusion, there is no question customers will slowly continue to adopt these new technologies.
Edge Computing and how it Transforms Enterprises
26:47Our guest on the podcast this week is Don Duet, President and COO at Vapor IO. We discuss the many definitions floating around for what edge computing is. Some call it fog computing or MEC (mobile edge computing). It is simple. It is a perspective that as the shift goes from wireless networks and person-to-person interaction to machine-to-machine interaction, underlying architectures must change along with that shift. For things like IoT and distributed data, they require reliability and speed at the source. Increased embeddedness from IoT to healthcare, finance to automotives requires lower latency and need solutions that are mass-scaled to meet distribution requirements. There are many use-cases for edge computing. Many often think first of Fitbits or Amazon Echos. However, it is transforming enterprises and changing entire business models based on the access to real-time data on the edge. It’s used in places like industrial IoT where sensors and sound collection takes place in factories. Managing these at the source helps solve scalability issues. Hospitals are transforming into mobile computing labs. As this technology enters the operating room, it needs to be 100% reliable and fast because lives are on the line. Last, autonomous vehicles and drones use edge computing. Decision-making is brought closer to the device, reducing steps of communication that can result in errors or latency and could cause accidents. Clearly, edge computing is the future in many budding industries.
Kubernetes Wins at Orchestration Engines, but Still Needs Backup
29:33Our guest on the podcast this week is Shannon Williams, Co-Founder and VP, Sales at Rancher Labs. We discuss how Kubernetes has won the war as a leader in orchestration. However, it is still not easy to use or maintain. We explore what organizations need to consider to build operational efficiencies around the technology. Kubernetes, Docker, and containers are very different from something like VMware and Amazon in terms of adoption. When new technology comes into an organization, they usually would quickly become an IT-led project. With containers, they are more like Puppet, Chef, and Ansible. They pop up in clusters around the organization, started directly by DevOps teams, developers, and users. A lot of times they already exist scattered around an organization and need a framework or logic to manage it all. That is what Rancher does, they act as a services platform to help manage many clusters. Key Considerations for Kubernetes We talk about the key considerations organizations need to think about when trying to deploy Kubernetes across multiple clusters across multi-cloud. Some key points are: Understanding expectations of scale Where will you be running it How to make it highly available Understanding your organization’s tolerance for failure What many don’t realize is that often organizations are running Kubernetes in conjunction with legacy technologies. If you keep up with cloud news, it can sometimes seem like everyone is using the latest technology. In reality, many organizations still use legacy technology like VMs and only make small incremental changes. The two biggest place Rancher runs containers is on VMware and on Amazon. In summary, Kubernetes may prolong the life of legacy technologies like VMware.
Kubernetes in Enterprises of all Sizes
26:04Our guest on the podcast this week is Brian Gracely, Director of Product Strategy at Red Hat. Brian discusses a small bank in Ohio and how they have a business use-case for Kubernetes. Because Kubernetes was built to manage Google-sized technology, it is surprising that there is a reason to apply it to a small brick and mortar bank just starting with web and mobile. What they noticed in the switch is that because customers get paid on Fridays and are more likely to check if they got paid on their personal mobile device, as soon as they launched mobile their Friday traffic soared. For just one fifth of business days, Fridays received ten times the amount of traffic as any other day. This unexpected spiky traffic pattern was a great use-case for Kubernetes. Kubernetes was built to deal with problems like that, even at small businesses. We also look at the state of the big three orchestration engines: Kubernetes, Mesos, and Swarm. Kubernetes and Mesos began as internal projects from larger companies. Mesos began in 2014 as a container scheduler Twitter was using to manage its own containers. They released it as open-source, so a community began to form around that. It focused on big data elements because of their applications to Twitter. To this day, Mesos is still preferred to run big data applications compared to the other two. Kubernetes began as Google’s Borg technology, used internally then released open-source. It was focused on the 80/20 type of use-cases such as batch use-cases and container use-cases. What happens when open-source solutions are released is that the community flocks to one and Kubernetes won more of the community than any solution. With a strong community, Kubernetes is better suited to work with many different types of applications. Last, Docker came out with Swarm to compete in the market. They keep things as simple as possible to get a few containers clustered together. Swarm has evolved to work mostly around Docker’s data center products. Tracking this industry over time reveals how containers have evolved to having use-cases in enterprises of all sizes.
RightScale’s 2017 State of the Cloud
27:33We discuss RightScale’s State of the Cloud report analyzing trends in the cloud. RightScale helps customers adopt cloud by helping them with a cloud management and optimization. This is the sixth year of the report so we can start to see trends over time now and there were a few interesting takeaways this year. In the report, RightScale asks two big questions for enterprises. First is about cloud strategy and what their intention is on cloud – to use private, public, or combinations of those. Second, they are asked about what they use today for private or public clouds. From a strategy point of view, people are still focused on multi-cloud with a special focus on hybrid cloud. In strategy, there was a shift away from private-only strategies. Fewer people were saying they plan to use only private cloud or multiple private clouds as their strategies. On adoption, there was a slight drop in people who are already using private cloud from 77% last year down to 72% this year. This may indicate companies who had tried to build their own private cloud with Openstack, and are now backing off from that strategy. The survey found that the average company leverages about four different cloud vendors. This a result of a combination of acquisitions of companies that use different cloud providers and a strategy to leverage different cloud providers. Rightscale asked people for a list of public and private clouds they are running applications on (focused on IaaS and PaaS, not SaaS) and whether they’re experimenting with particular public and private clouds. They found that among people that are using at least one public cloud, they’re running applications in 1.8 public clouds. They are typically experimenting with another 1.8 clouds. Even if they are not using one of the big cloud vendors, they are often at least experimenting with it. The ones that are adopting private cloud are reporting about 2.3 different private clouds. For top challenges this year there was a three-way tie between security, spend and skills (access to skilled resources). Last year skills was highest and it has dropped a bit this year. The people in IT who are concerned about security has been declining each year. Among enterprises, in 2017 over 35% rated cloud security as a significant challenge, and six years ago that number was about 10% higher. We have now reached a tipping point where people realize that when done right, cloud can be as secure if not more secure than a traditional data center. As people adopt public cloud, the cost has been increasing and companies are starting to realize they are inefficient with their spend. On average companies believe they are wasting 30% of their cloud spend. RightScale has found that 30-45% or more is typically what companies are wasting on their cloud spend. The survey found that the more mature a cloud instance is, the more important spend becomes. This year Docker has moved into first place in the list of tools RightScale researches, and while all tools had an increase in usage, Chef and Puppet had a decrease in usage. The survey specifically focuses on configuration management tools and container tools. Docker usage moved from 13% in 2015 to 27% in 2016 to 35% this year, while Chef and Puppet each dropped about 4% this year. The other big increase seen this year was in Kubernetes, which doubled from 7% last year to 14% this year and seems to be in the lead for scheduling and orchestration tools. There is an early trend RightScale noticed that people are starting to use Docker to take advantage of the temporary instances from the cloud providers such as AWS Spot or Google Preemptibles. For people looking to use those, which can mean 70-90% savings on demand, they need the ability to be very portable when they lose their temporary instances, so using Docker along with a container as a service can be helpful in saving those costs. We look at predictions for next year’s State of the Cloud report. Private cloud will likely continue to be under pressure, though we may see a slight uptick with VMware on AWS. It is likely that Docker will continue to grow and that the cost of the cloud will continue to be an ongoing challenge for enterprises.
The DevOps Handbook for Unicorns and Horses
23:51Our guest on the podcast this week is Gene Kim, DevOps expert and author at "The Phoenix Project" and "The DevOps Handbook." We discuss “The DevOps Handbook”, which was started over five and a half years ago and released in October 2016. Gene co-wrote the book with Jez Humble, Patrick Debois, and John Willis. The book includes over 48 case studies that range from unicorns like Google, Amazon, Facebook, but also horses like Nordstrom, Target, and Capital One. Many of the case studies came from the DevOps Enterprise Summit. The book has discussions of both greenfield and brownfield deployments, even touching on mainframes. The most common question enterprises ask about DevOps is: Where do we start? “The DevOps Handbook” starts with a chapter on starting with DevOps by picking the right value stream. The research started by looking at where successful unicorns and horses started with their DevOps initiatives. They also look at the failed DevOps initiatives to learn what to avoid. Failures can be categorized in two ways: starting too small or starting too big. Initiatives that start too small often start with a simple Chef or Puppet project end up looking like like more of a hobby. When they finish the project, they haven’t actually proven anything and the project gets easily dismissed. Initiatives that start too large often choose something too critical to the operations of the organization, and is unforgiving of mistakes. The most successful journeys start with something that creates a material contribution to the organization, but is small enough that it does not get shut down early for small mistakes. Their studies found that one out of three leaders who were starting these transformations were being promoted for their contributions. There is no one answer for how to change an organization’s culture for DevOps, but there is a prescriptive set of guidelines. The technical practices do not change such as version control, continuous testing, continuous integration, automated deployments, proactive monitoring of the production environment, security integrated into every step. What is different is where these initiatives start from. Some come from the Director ofr Operations, a Chief Architect, or even Director of Development. This transformation starts from different people and teams at different organizations. There are many different ways to reach a great DevOps practice. The three guiding principles of DevOps are: Flow : Maximizing the flow of work and minimizing the lead-time Feedback: Creating check-ins and the equivalent of being able to stop the assembly line Culture: Fostering a culture of continuous experimentation and learning There is a myth that with DevOps you can’t have any central control. We discuss differences in self-service teams like Netflix and Amazon to function-based teams used at Google and Disney. We look at Etsy’s liason model assigning ops engineers to various service and product teams. Many companies do well with the technical aspects of DevOps, but struggle with the culture changes it requires. Ten years from now, it will be about creating learning organizations and the command and control model will not be effective. It won’t be about who caused a problem or who to blame, but will be about creating a culture of learning for a successful team.
Security and DevOps are Never 100% Done
34:19Our guest on the podcast this week is Jim Ford, Chief Strategic Architect at ADP. We discuss Docker adoption, DevOps, and security for a company as big as ADP with over 630,000 customers, 35 million users in 100 countries, and 55 million stored social security numbers. We see how DevOps processes and security is never a completed initiative and can always be improved. We also see how DevOps at ADP varies significantly across projects and seems to mold to the team using it.
Hybrid Cloud is Not a Choice, It’s a Realization
36:07Our guest on the podcast this week is Rob Hirschfeld, Founder and CEO at RackN. We discuss the differences between Docker Swarm and Kubernetes as well as the debate about how many companies use containers the right way. We also look at hybrid and how it can be easy to lift a container from Amazon and move it to Google, but there will still be key differences in networking and storage between the two that make the shift more complex. Last, we talk about the pros and cons of going all in on one vendor.