Philadelphia, Pennsylvania
April 2–4, 2019
Click Here For Information & Registration

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Cloud Foundry for Operators [clear filter]
Wednesday, April 3


Introduction to Cloud Foundry for Operators Track - Marco Voelz, SAP & James Hunt, S&W
avatar for Marco Voelz

Marco Voelz

Software Engineer, SAP
Marco is a software developer at SAP working as PM in the BOSH Team Europe. He is a committer in BOSH and spends most of his time with monitoring, automation, and nitty-gritty infrastructure details.
avatar for James Hunt

James Hunt

Chief Architect, S&W
James Hunt is the Chief Architect and Technical Evangelist at Stark & Wayne, a leading Cloud Foundry consultancy. He has authored dozens of Open Source projects, including: SHIELD, Safe, Genesis, Spruce, and more.

Wednesday April 3, 2019 11:10am - 11:20am
Room 122A


Double, Double Toil and Trouble; SRE Transformation through Automation and Collaboration - Bozhidar Lenchov, Corey Innis & Ryan Collins, Pivotal
As security, reliability and continuous integration become more of a concern for organizations of all sizes, DevOps teams have to come up with ways to support these increasing workload demands. Operating a Platform as a Service, managed by a team over multiple timezones, presents additional technical and communicational challenges which we've tried to identify and address.

We all have many tedious chores related to maintaining on our production environment(s). So, we automate the toil; now what do we do? Oh, we can start collaborating with other teams to help reduce their toil! Now, we've enabled self-service, so now what's next? Oh, we can share these learnings via workshops because it turns out that dev teams also benefit from this knowledge base and set of practices. That is, we scale!

In this presentation Bozhidar, Corey and Ryan from the Pivotal CloudOps teams in Toronto and San Francisco will outline some of the automation and practices used in their day-to-day operations. Those include detection of abusive workloads, automating platform updates and backups and auditing expiring credentials and potentially-malicious clients to prevent downtime and remain within SLI/SLOs. We'll also discuss practices around handling interrupts and incidents, working remotely, enabling teams to self-service requests and sharing context within the team and to the broader organization.

avatar for Ryan Collins

Ryan Collins

Manager, Software Engineering, Pivotal
avatar for Bozhidar Lenchov

Bozhidar Lenchov

Software Engineer, Pivotal
Bozhidar is a Senior Software Engineer at Pivotal, working in the Toronto, Canada office. They are currently part of the team managing Pivotal Web Services, but have previously spent time contributing on the BOSH Core team. Prior to Cloud Foundry, they spent 4 years working on and... Read More →

Corey Innis

Software Engineer, Pivotal
Corey is a Staff Software Engineer at Pivotal in San Francisco. They first joined Pivotal in 2007, and have found a passion for helping transform teams and organizations through Agile, Lean, Balanced Team and DevOps/SRE practices (and more).

Wednesday April 3, 2019 11:20am - 11:50am
Room 122A
  • Experience Level Any


Air-Gapping: A Moat for the 21st Century - Michael Ferris, Stark & Wayne
The best way to ensure that an unauthorized, external bad actor cannot take advantage of your platform is to make sure they cannot reach it in the first place. For this reason, many organizations choose to run their PaaS in an on-prem, air-gapped environment, meaning there is no path between the public internet and your PaaS, either inbound or outbound. However, air-gapping a PaaS without using the proper processes and tools can drastically slow down your Operators and undermine the benefits that the organization adopted the PaaS for in the first place. In this talk, Mike will outline the implications of running your PaaS in an air-gapped environment, including changes to the installation process, the running of apps, and the ongoing maintenance of your air-gapped platform. Vince White from Agile Defense will share his war stories with you about the most common pitfalls of running a PaaS in an air-gapped environment and how to avoid them.


Michael Ferris

Cloud Engineer, Stark & Wayne
Mike Ferris is a Cloud Engineer at Stark & Wayne. He brings years of experience operating production-grade PaaS. Mike also helps clients install, maintain, and automate their platforms. Over the past 18 months, Mike has consulted for Pivotal, helping them roll out and operate their... Read More →

Wednesday April 3, 2019 12:00pm - 12:30pm
Room 122A
  • Experience Level Any


Debugging the Routing Tier - Angela Chin & Nitya Dhanushkodi, Pivotal
As the entrypoint to Cloud Foundry, routing-tier related errors come from a variety of places. While some of these errors are indications of actual bugs in Cloud Foundry source code, there can be a number of other root causes-- misbehaving applications, misconfigured load balancers, and infrastructure issues, just to name a few. In order to isolate and debug these problems, it becomes important to know which data to collect, how to collect it, and what in the data can indicate the root cause.

In this talk, Angela and Nitya will cover what information is useful to collect and what to look for in the data to systematically eliminate possible causes of the problem. Drawing on their experiences as members of the CF Routing team, they will go through issues that operators have seen in the past, and how they use tools such as routing logs, pprof, flamegraphs, and wireshark to debug, isolate, and work around these issues.

avatar for Angela Chin

Angela Chin

Senior Software Engineer, Pivotal
Angela is a software engineer at Pivotal, currently working on all things networking and service mesh related. She has contributed to open source Cloud Foundry, primarily in areas related to networking and routing, and also previously worked on improving the Day 2 experience of Kubernetes... Read More →
avatar for Nitya Dhanushkodi

Nitya Dhanushkodi

Software Engineer, Pivotal
Nitya Dhanushkodi is a software engineer at Pivotal, working on the Cloud Foundry Networking team. She is currently working on integrating the Istio service mesh with Cloud Foundry. She has also worked on the CF MySQL and Credhub teams prior to Networking. Nitya has given a previous... Read More →

Wednesday April 3, 2019 2:15pm - 2:45pm
Room 122A


Extending Cloud Foundry at Scale - Addressing the Non-12-factor App Gap with CFCR - Colby Johnston & Brett Mayer, Comcast
For many years Comcast has been successfully running and managing Cloud Foundry for stateless applications and micro-services at scale. Over time, we have had to address the requirements of our non-12-factor apps as well. We have chosen Container Runtime (CFCR) platform for these stateful applications by deploying multi-site / multi-cloud environments. As an extension of Cloud Foundry Application Runtime, it allows customers to do self service deployments and scaling of application stacks and make use of additional features such as service catalog/market place, eternal DNS and persistent storage. Join us to learn and provide feedback on how we have leveraged Bosh and Concourse to efficiently and predictably deploy and manage CFCR at scale.

avatar for Colby Johnston

Colby Johnston

Senior Cloud Engineer, Comcast
I've been working in IT since 1996 working in various systems engineering roles. The previous 3 years, I've been part of the Comcast Cloud Engineering team working on Cloud platforms including CF, AWS, SDS, Kubernetes, and now CFCR. My primary role has been cloud platform engineering... Read More →

Brett Mayer

Senior Cloud Engineer, Comcast
Senior Cloud Engineer at Comcast responsible for engineering, deployment, management, of cloud platforms.Speaking experience has been with internal Comcast teams, organizations, and customers.

Wednesday April 3, 2019 2:55pm - 3:25pm
Room 122A
  • Experience Level Any


Preventing Doomsday: x509 Certificates in Cloud Foundry - Tom Mitchell & David Dobmeier, Stark & Wayne
This talk covers the use and rotation of TLS certificates in distributed systems to enhance security and the ongoing monitoring of certificate expiry required to ensure uptime. We take a look at Cloud Foundry as a case study on the use of mutual TLS between networked components, where the number of required certificates can grow large and non-trivial to manage. We share some of our own stories with Cloud Foundry and BOSH certificates expiring, the effect this can have on the system as a whole, and methods we’ve used to avoid the downtime this causes.

avatar for David Dobmeier

David Dobmeier

Cloud Engineer, Stark & Wayne
David Dobmeier is a Cloud Engineer at Stark & Wayne. He has seven years of experience including Systems/Network Administration to DevOps, Software Development, and Cloud Infrastructure. In the past few years, he has focused on large-scale infrastructure and software systems including... Read More →

Tom Mitchell

Stark & Wayne
Tom Mitchell is a Cloud Engineer at Stark & Wayne. He spends his days consulting on Cloud Foundry and BOSH, and creating tools to make life easier for other people working in the domain. He types in Dvorak and makes sure to tell everybody.

Wednesday April 3, 2019 3:35pm - 4:05pm
Room 122A
  • Experience Level Any


Tired of Onboarding Toil? - Caleb Washburn, Pivotal
Come learn about how you can leverage cf-mgmt to reduce the toil of onboarding and managing configuration of your Cloud Foundry deployments.

- Demonstrate how to get started using cf-mgmt to manage your existing environments in minutes

- Demonstrate how to extend your configuration going forward on a day to day basis and how you can distribute this workflow to your consumers in a self-service way.

- Demonstrate how to use ldap/active directory groups to manage space/org roles

- and more...

avatar for Caleb Washburn

Caleb Washburn

Director, PCF Solutions, Pivotal
Director at Pivotal. Specializing in large scale automation across enterprise operations. Have successfully overhauled several of our F500 software platform engineering teams, written tooling in and around operational aspects of managing a distributed platform, and implemented fully... Read More →

Wednesday April 3, 2019 4:35pm - 5:05pm
Room 122A


Bulk Application Migration Between Cloud Foundries - Michael Brodhead, & Patrick Jones, Stark & Wayne
You have 30 days to migrate hundreds of applications from one Cloud Foundry environment to another. You don’t know what any of the apps are, nor how they work. How will you take inventory of what is running? How will you move these apps to the new site? How do you know whether they have been migrated properly? Believe it or not, this is all manageable. Learn what to do from a team which has done the job many times and lived to tell about it. Ready? Go.

avatar for Michael Brodhead

Michael Brodhead

Cloud Security Architect, Stark & Wayne
Michael Brodhead, MKB, is a Cloud Security Architect at Stark & Wayne. He has over twenty years of experience in information security and software development. MKB has worked for financial giants, tiny startups, and everything in between. At Stark & Wayne, he helps clients secure... Read More →
avatar for Patrick Jones

Patrick Jones

Cloud Engineer, Stark & Wayne
Patrick Jones is a Cloud Engineer at Stark & Wayne. Patrick spends most of his time working on open source projects like SHIELD and Genesis as well as building and maintaining large-scale open source Cloud Foundry environments. Patrick has also spoken at and taught hands-on labs at... Read More →

Wednesday April 3, 2019 5:15pm - 5:45pm
Room 122A
Thursday, April 4


A Native Kube Operator Tailored for Cloud Foundry - Troy Topnik, SUSE & Enrique Encalada, IBM
In this talk the audience will learn about the cf-operator, a project that's part of the CF Containerization proposal.
We'll show what this operator can do and how Cloud Foundry deployments can be managed with it.

We'll also delve deeper, and talk about implementation techniques, Kubernetes Controllers and Custom Resources.

This will be a great opportunity to learn about how Cloud Foundry can work flawlessly on top of Kube.

avatar for Troy Topnik

Troy Topnik

Senior Product Manager, SUSE
Troy is a Senior Product Manager responsible for SUSE Cloud Application Platform. He began working with Cloud Foundry shortly after its open source debut in 2011, and has been a technical writer, instructor, and product manager with the ActiveState and HPE Helion Stackato teams.

Enrique Encalada

Software Engineer, IBM Germany
Enrique Encalada is a software developer currently working on the Cloud Computing field, with a strong focus on Kubernetes and CloudFoundry. Enrique likes to work with different technologies, try to understand them and build something upon that. Part of Enrique´s spare time goes... Read More →

Thursday April 4, 2019 11:05am - 11:35am
Room 122A


Parting the Clouds: Tracing Cloud Foundry for Improved Performance and Availability - Isobel Redelmeier, LightStep
Pushing a new app is fast - except when something times out. Maybe scaling takes too long. And, of course, you need more instances of some apps than you'd expect, or you just can't trust them to perform as well as you'd like.

How well do you really know what's going on in your Cloud Foundry? How much time do you spend ensuring availability? How much do you spend keeping excess nodes around, "just in case"?

Whether you're an operator trying to make sure everything is running smoothly for your developers, or you're a contributor working on Cloud Foundry itself and want to gain better insight into how your components actually behave, you'll learn how distributed tracing can keep you better informed about how your platform is actually behaving so that you can spend less time and money ensuring everything is performing to your standards.

The talk will cover the basics of what tracing itself is, before a deep dive into some examples of how it can help Cloud Foundry operation and development.

avatar for Isobel Redelmeier

Isobel Redelmeier

Software Engineer, LightStep
Isobel Redelmeier works on observability at LightStep, where she focuses on OpenTelemetry and the overall goal of making the software world more debuggable. She's spoken at conferences such as Velocity, OSCON, RustConf, and gRPConf on a variety of topics. Ask her about observability... Read More →

Thursday April 4, 2019 11:45am - 12:15pm
Room 122A
  • Experience Level Any


Diego vs Eirini: How to Choose the Best Fitting Container Scheduler - Julian Skupnjak, IBM R&D & Georgi Dankov, SAP
Now that Eirini is an official incubation project, operators can choose between Kubernetes and Diego as their container scheduler in Cloud Foundry. But how should you - as an operator - make that choice? What are the advantages and disadvantages of each?

Diego and Kubernetes are both highly scalable, popular scheduling technologies. One has been built into Cloud Foundry for a long time. One, via Eirini, has recently become available to Cloud Foundry operators. Comparing them, one will find many similarities, but also more than a few differences. This brings up the question: when am I supposed to choose one vs the other?

In this talk, we will show the similarities between the two container orchestrators, and the key differences. We’ll give Cloud Foundry customers and operators a better understanding of in which scenarios it is useful to have Diego as container orchestrator, and in which scenarios it might be more useful to use Kubernetes.

avatar for Julian Skupnjak

Julian Skupnjak

Software Engineer, IBM
Julian Skupnjak (Herr Julz) is a Software Engineer at IBM Cloud and part of Cloud Foundry's Eirini development team. During his career at IBM Cloud he worked on IBM's Cloud Foundry production system in different roles, but mostly as DevOps engineer at the internal delivery team. During... Read More →

Georgi Dankov

Software Engineer, SAP
Georgi Dankov works as a Software Engineer at SAP in Bulgaria. Previously he has worked on internal services running on SAP Cloud Platform’s Neo environment. Currently he is part of the team, working on the Cloud Foundry incubating project Eirini - responsible for enabling pluggable... Read More →

Thursday April 4, 2019 2:00pm - 2:30pm
Room 122A


Smelling Smoke? How to Design a Disaster Recovery Strategy for Cloud Foundry - George Lestaris & Chunyi Lyu, Pivotal
Cloud Foundry is a platform that can run on any IaaS. It is a distributed system composed mostly of stateless components, in true cloud-native fashion, but state has to, eventually, live somewhere. Disaster recovery (DR) is usually a non-functional operational readiness requirement but it is not always well-understood. Platform Engineers need to have a trusted DR plan as it’s crucial to both their business continuity strategy and to the developers of the platform.

This talk is split into two sections. The first part will be about the theory of DR for distributed systems and how this theory applies to Cloud Foundry. DR comes with a set of tradeoffs that platform engineers need to incorporate into their platform architecture. In the second part, Chunyi and George will be talking about BOSH Backup a Restore (BBR). A tool that can be part of a DR plan when Cloud Foundry is deployed using BOSH. They will discuss patterns of multi-site setups and the tradeoffs platform engineers can make in order to minimise the impact of backups.


Chunyi Lyu

Software Engineer, Pivotal
Chunyi Lyu is a software engineer at Pivotal working on the Platform Recovery(BBR) team. She has extensive experience on the open source Cloud Foundry and BOSH, and has previously spoken at CF summit 2018.
avatar for George Lestaris

George Lestaris

Product Manager, Pivotal
George Lestaris is the product manager for the Cloud Foundry BBR team, working at Pivotal in London. Prior to this, he led the CFCR and GrootFS projects. Before Pivotal, George spent time with high throughput computing and cloud computing research. He has given talks before in CF... Read More →

Thursday April 4, 2019 2:40pm - 3:10pm
Room 122A


Ways to Monitor your BOSH Director - Beyhan Veli, SAP SE
The BOSH Team Europe at SAP operates BOSH as a service for operators and the Cloud Foundry on-demand service brokers. Like every production service, the expectations to BOSH's performance and availability have increased dramatically.

Motivated by concrete issues in our production BOSH services, we have continuously worked on identifying the useful metrics and making them available to monitoring tools. No one likes to be paged and that is why it was a crucial task for us to improve the BOSH monitoring.

In this talk we share some of our operation war stories and the metrics which we identified to prevent these situations as well as some ways to expose those metrics to monitoring tools. Moreover, we will give an update on all the improvements implemented for BOSH to make a more sophisticated monitoring possible.


Beyhan Veli

Software Engineer, SAP SE
Beyhan is a developer in the BOSH Team Europe. He is a committer on the BOSH project for over 3 years now. Before that he was working on different topics for the SAP’s Cloud Platform.

Thursday April 4, 2019 3:40pm - 4:10pm
Room 122A


Cloud Foundry Availability - Julian Fischer, anynines GmbH
The definition and measurement of availability is an integral part of any system’s operational concept. This talk guides through the definition of availability of a Cloud Foundry based platform. This involves the question on what subsystems are relevant to the definition and how they contribute to the availability metric in the end. Once, a definition has been found a brief the measurement of availability is discussed. With a well defined availability metric at hand, the journey of uptime optimization can be started and will be briefly elaborated. By the end of this journey, you will have learned about defining, measuring and improving the availability of not only Cloud Foundry but complex distributed systems in general and know why maintenance windows become obsolete.

avatar for Julian Fischer

Julian Fischer

CEO, anynines GmbH
Julian Fischer, CEO of anynines, has dedicated his career to the improvement and automation of software operations. In more than fifteen years, he has built several application platforms with various open source automation tools. His latest passions are Data Service Automation, Cloud... Read More →

Thursday April 4, 2019 4:20pm - 4:50pm
Room 122A