Life at Eclipse

Musings on the Eclipse Foundation, the community and the ecosystem

Archive for the ‘Open Source’ Category

Eclipse ioFog: Evolving Toward Native Kubernetes Orchestration at the Edge

leave a comment »

With the proliferation of AI, autonomous vehicles, 5G, IoT, and other industrial use cases that require lightning-fast data processing, edge computing has emerged over the past few years as a way to address the limitations of existing cloud architectures to process information and deliver services at the “IoT edge”. Instead of backhauling data to the centralized cloud, edge computing brings computational power closer to data sources to support near real-time insights and local actions while reducing network bandwidth and storage requirements.

According to Gartner, 75% of enterprise-generated data will be created and processed outside a traditional centralized data center or cloud by 2025. While the problems at the IoT edge — connectivity, manageability, scalability, reliability, security — are being solved as point solutions by enterprises and ecosystem players, there is a need for a foundational industry-wide standard for managing distributed IoT workloads. Time and again, open source has been proven to be the best way to deliver complex platform software with industrial scale collaboration.

Enter Kubernetes, the de facto standard for orchestrating containers and the applications running inside them. Kubernetes has massive potential for handling IoT workloads on the edge by providing a common control plane across hybrid cloud and edge environments to simplify management and operations. Within the Kubernetes IoT Edge Working Group, members of the Kubernetes and Eclipse communities are collaborating with leading technology innovators to extend Kubernetes to support dynamically, securely, and remotely managing edge nodes.

A great example of this collaboration is Eclipse ioFog, a universal Edge Compute Platform which offers a standardized way to develop and remotely deploy secure microservices to edge computing devices. ioFog can be installed on any hardware running Linux and provides a universal runtime for microservices to dynamically run on the edge. Companies in different vertical markets such as retail, automotive, oil and gas, telco, and healthcare are using ioFog to turn any compute device into an edge software platform.

The Eclipse Foundation is excited to support today’s announcement of the initial availability of ioFog features that make any Kubernetes distribution edge-aware. With this latest release, developers are able to extend Kubernetes to easily deploy, secure, and manage edge computing networks supporting applications such as advanced AI and machine learning algorithms.

Farah Papaioannou, co-founder and president of Edgeworx, explains the significance of the release this way:

“ioFog is a proven platform at the edge. With this release, we build on native Kubernetes, seamlessly extending it to the edge…We do this based on things that actually matter at the edge, such as latency, location or resources. We are delivering today a full cloud-to-edge solution, that’s 100-percent open source and works with any Kubernetes flavors and distros.”

These native Kubernetes enhancements are in the process of being contributed to the Eclipse ioFog open source project. We are proud to support the collective efforts of the Eclipse community and the Kubernetes ecosystem to help developers deploy, manage, and orchestrate applications and microservices from cloud to edge in a simple and secure manner.

For more information about ioFog, get started by using this link to install and set up your production ioFog environment. If you have questions or want to connect with other people involved in this platform, join the ioFog community and the mailing list.

Written by Mike Milinkovich

June 23, 2019 at 4:46 pm

Posted in Foundation, Open Source

Update on Jakarta EE Rights to Java Trademarks

Summary of progress to date and implications of the agreement between Eclipse and Oracle on Jakarta EE and use of Java trademarks and the javax namespace.

Introduction

The migration of Java EE to the Eclipse Foundation has been an enormous effort on behalf of the Eclipse Foundation staff and the many contributors, committers, members, and stakeholders that are participating.

This post was reviewed and represents the consensus view of the Jakarta EE Steering Committee.

Earlier this year GlassFish 5.1 was certified at Eclipse using the Java EE 8 TCK at the Eclipse Foundation. Since then under the direction of the working group the community established a Jakarta EE Specification Process (JESP) for the evolution of the Jakarta EE specs at the Eclipse Foundation. Specification projects are being created for all of the Jakarta EE specifications. The TCK process is being refined for Jakarta EE in concert with the new Jakarta EE compatibility logo. This is all being done in support of the Jakarta EE 8 release.

Progress to Date

The Jakarta community has been busy.

  • Oracle contributed GlassFish and the Java EE APIs and TCKs to Jakarta EE.
  • The Jakarta EE Working Group was formed and supporting committees to provide governance to the community and facilitate collaboration.
  • Eclipse GlassFish was certified with the Java EE TCK at the Eclipse Foundation.
  • The Eclipse Foundation Specification Process was created, and customization created and approved for the Jakarta EE Specification Process.
  • Specification projects are being created and work is underway now within the community to deliver the Jakarta EE 8 release later this year.
  • There is an initiative underway for Oracle, IBM, Red Hat and other members of the JCP to contribute their specification documents created at the JCP to Jakarta.

It had been the mutual intention of the Eclipse Foundation and Oracle to agree to terms that would allow the evolution of the javax package namespace in Jakarta EE specifications.   Unfortunately, following many months of good-faith negotiations, the Eclipse Foundation and Oracle have been unable to agree on terms of an agreement for the Eclipse Foundation community to modify the javax package namespace or to use the Java trademarks currently used in Java EE specifications.   Instead, Eclipse and Oracle have agreed that the javax package namespace cannot be evolved by the Jakarta EE community. As well, Java trademarks such as the existing specification names cannot be used by Jakarta EE specifications.  Because of the complexity and confidential nature of our negotiations, the Eclipse Foundation and Oracle have also agreed that we will not attempt to characterize here what has resulted in this outcome. It is the best outcome we could mutually achieve for the community. Some additional context is provided in the Eclipse Foundation Board and Jakarta EE Steering Committee meeting minutes.

What restrictions does this outcome impose on the Eclipse community?

Oracle’s Java trademarks are the property of Oracle and the Eclipse Foundation has no rights to use them.   The implications of this are as follows:

  1. The javax package namespace may be used within Jakarta EE specifications but may be used “as is” only.  No modification to the javax package namespace is permitted within Jakarta EE component specifications. Jakarta EE specifications that continue to use the javax package namespace must remain TCK compatible with the corresponding Java EE specifications.
  2. Jakarta EE component specifications using the javax package namespace may be omitted entirely from future Jakarta EE Platform specifications.
  3. Specification names must be changed from a “Java EE” naming convention to a “Jakarta EE” naming convention.  This includes acronyms such as EJB, JPA or JAX-RS.

In addition to the above, any specifications which use the javax namespace will continue to carry the certification and container requirements which Java EE has had in the past. I.e., implementations which claim compliance with any version of the Jakarta EE specifications using the javax namespace must test on and distribute containers which embed certified Java SE implementations licensed by Oracle. These restrictions do not apply to Jakarta EE specifications which do not utilize javax, including future revisions of the platform specifications which eliminate javax.

Note that the ratified Jakarta EE specifications will be available under a different license (the Eclipse Foundation Specification License). This is the reason why the Eclipse Foundation is currently asking the community to update their contributor and committer agreements.

What is next for the Jakarta EE Working Group?

The Jakarta EE Working Group including Oracle will continue to do what they set out to do: namely, move Java EE to the Eclipse Foundation. The group remains committed to creating a Jakarta EE 8 specification, logoed under the Eclipse Foundation’s Jakarta trademark. Further, the group is also committed to future versions of the Jakarta EE specifications that deliver on the original promises of innovation and evolution in cloud-native Java.

What does it mean for Jakarta EE to not modify the javax package namespace?

The Agreement does allow for modification and evolution under a new namespace, such as jakarta. It is expected that all future evolution and innovation will not use the javax namespace.

What happens to the Jakarta EE brand?

The Jakarta EE compatibility and the Jakarta EE member logos have both been decided on by the community and published. Work is underway to deliver the branding usage guidelines and supporting trademark license agreement. We expect to see the usage of these brands later this year.

Will there be a Jakarta EE 8?

Yes. The community is working hard to deliver the Jakarta EE 8 release, which is Java EE 8 delivered from Eclipse. We expect that many application servers will certify as Jakarta EE 8 compatible.

What happens beyond Jakarta EE 8?

The guiding principle for Jakarta EE 9 will be to maximize compatibility with Jakarta EE 8 for future versions without stifling innovation.  This will most likely involve two key topics: migration of some or all of the Jakarta EE specification source to a new namespace for future evolution; means to provide backwards compatibility with javax at a binary level, allowing old applications to run on Jakarta 9 implementations with some form of build or runtime tooling.

So while there will be a point in time where future versions of specifications will have to go through a source code incompatible change with respect to the previous javax based packages, this will be a straightforward transformation.

Further plans are being evolved by the Jakarta EE community via the Jakarta EE Platform Project. Your comments and participation are encouraged.

What does this mean for the EE4J projects?

The specification projects need to be renamed to not use Oracle’s Java trademarks. The Jakarta community gets to decide on the new names. This is an opportunity to tighten up the naming and get a better level of consistency. The future of Eclipse Enterprise for Java (EE4J) projects will be determined by the community who participate in those specifications and open source projects.

What is next for Eclipse GlassFish?

Work is underway at the Eclipse GlassFish project running the Jakarta EE 8 TCK and being ready to support its role in the TCK for interop testing. The future of Eclipse GlassFish will be determined by the community who participate in the project.

How will specs be updated beyond Jakarta EE 8?

Jakarta EE specifications will be updated according to the approved Jakarta EE Specification Process (JESP). The actual enhancements will be determined by the community who participate in those specification projects.

Written by Mike Milinkovich

May 3, 2019 at 6:00 am

The 2019 IoT Developer Survey Results are Live

After months of hard work, the 2019 IoT Developer Survey results are live today. This year marks the fifth year the Eclipse IoT Working Group has asked the global IoT developer community to share their perceptions, requirements, and priorities. The survey has proven to be an influential assessment of the IoT market as viewed from the development front lines. Access the full findings of the 2019 IoT Developer Survey here.

Over 1,700 individuals took the survey between February and March 2019. Just like in previous years (see results from 2018, 2017, and earlier here), Eclipse IoT collaborated with key IoT ecosystem players like Bosch and Red Hat to maximize the reach of the survey.

The key findings this year include the following:

  • IoT drives real-world, commercial outcomes today. 65% of respondents are currently working on IoT projects professionally or will be in the next 18 months.
  • IoT developers mostly use C, C++, Java, JavaScript, and Python
  • AWS, Azure, and GCP are the leading IoT cloud platforms
  • Top three industry focus areas remain the same as last year: IoT Platforms, Home Automation, and Industrial Automation / IIoT.
  • MQTT remains the dominant IoT communication protocol leveraged by developers
  • The Eclipse Desktop IDE is the leading IDE for building IoT applications

IoT gets real(er)

Consistent with our findings last year, two-thirds of the respondents to our survey develop and deploy IoT solutions today or will be doing so within 18 months. This continued focus on building and deploying real world solutions is reflected in the increases in developers’ focus on performance, connectivity, and standards shown in the survey.

C and Java dominate

C won out as the programming language of choice for constrained devices, while Java was most popular for gateways/edge nodes and IoT cloud. Neither of those findings are surprising. C and C++ have long been the languages of choice for small embedded systems where minimizing memory space and power consumption, and maximizing processor utilization are key. Java is the dominant language and platform where the memory and processing resources are larger, and the complexity of the systems are greater. In particular, Java is the language of choice for most cloud infrastructure projects, so seeing it lead in IoT cloud is consistent with that.

AWS, Azure, and Google hold on to the lead

As expected AWS, Azure, and GCP maintain their status as the leading IoT cloud platforms. The list of three and their rankings are entirely consistent with their relative weights in the cloud computing marketplace as a whole, so there is no surprise to see this reflected in our results.

A continued focus on platforms, home automation, and IIoT

IoT Platforms (34%), Home Automation (27%), and Industrial Automation / IIoT (26%) were the respondents’ three most common industry focus areas. These areas are likely to continue to be key targets for IoT developer activity.  The fact that IoT Platforms is consistently year after year the number one focus for IoT developers is interesting. It implies that enterprises and industrials are putting resources into building their own IoT platforms for use by their companies. To me this suggests that industrial IoT is going to be a huge opportunity for hybrid cloud, as companies build and run IoT solutions on-premise based using modern, open technologies.

Security is (still) top of mind

Security is still the top concern for IoT developers. Communication Security (38%) Data Encryption (38%), and JSON Web Tokens (JWTs) (26%) were the top three security technologies cited in the survey, with virtualization also starting to play a stronger role in IoT Security.

MQTT is still the dominant IoT communication protocol

HTTP (49%), MQTT (42%), and Websockets (26%) were the top three communications protocols used by IoT developers. The growth in MQTT adoption over the past seven or eight years has been phenomenal, and I like to think that the Eclipse IoT community with its Eclipse Paho and Eclipse Mosquitto projects had a small part to play in that. Having robust open source implementations available has certainly been part of the MQTT’s success. Looking forward that main challenge we see for further MQTT adoption is the lack of interoperability built into the protocol. While MQTT is a great lightweight, low latency protocol it does not provide any guidance on the topic structures and payload definitions used by any device or application. This means that no two teams using MQTT would expect to have their systems be able to reliably exchange data. The Eclipse Tahu project defines the Sparkplug protocol — created by Arlen Nipper one of the co-inventors of MQTT itself. Sparkplug defines the topic structures and payload definitions necessary for out-of-the-box interoperability of SCADA systems. We are hopeful that Sparkplug could spur MQTT to even greater adoption in industrial IoT use cases.

The Eclipse Desktop IDE is the leading IoT IDE

45% of respondents use the Eclipse Desktop IDE. It is not at all surprising that the Eclipse IDE has a strong franchise with IoT developers, given the dominance of C and Java. The Eclipse CDT project has long been hugely important in the embedded software space. The past decade CDT has been used by virtually every chip, SOC, and RTOS company as the basis for their toolset. Those developer solutions also typically use additional tools such as the Target Management Framework, and Remote Systems Explorer that were specifically designed with the embedded developer in mind. That, coupled with the Eclipse IDE’s broad use amongst professional Java developers makes its leadership in IoT clear.

In addition, close to 10% also use Eclipse Che, our community’s next generation cloud-based tooling solution. It really seems part of the future of IoT is in the cloud, one way or another.

Access the full findings of the 2019 IoT Developer Survey here.

Thanks to everyone who took the time to fill out this survey, and thanks again our Eclipse IoT members for their help with the promotion.

We are very interested in hearing your thoughts and feedback about this year’s findings. And, of course, we are always open to suggestions on how to improve the survey in the future!

Written by Mike Milinkovich

April 17, 2019 at 7:00 am

Posted in Foundation, Open Source

New Eclipse Foundation Committer and Contributor Agreements

Note: to see redline versions of the changes to the documents discussed below, please visit this contribution and committer agreements page.

Over my almost 15 years of sharing updates about what’s going on at Eclipse, some blogs are more important than others.  This one is important as it requires action by our members, committers, and contributors!  There is a lot of ground to cover explaining what’s going on and why we’re changing things, so please forgive me for a longer than normal post.

tl;dr.  The Eclipse Foundation is starting to develop specifications. First for Jakarta EE, but soon for other areas as well. We want to make it clear that contributions to our open source projects may someday be used to create a specification, because we believe in code-first innovation. We also believe that if you’re contributing to open source, you want your contributions to be used for open purposes, including specs.

We are updating our standard contributor and committer agreements, and we will be requiring all our committers and contributors, as well as those members who have member committer agreements,  to re-sign their agreement with us.

To make this happen, we will be reaching out to everyone who needs to re-sign.  You don’t have to do anything yet – just be aware the change is coming, and please act when we do make contact with you.

First, a bit of background.  All contributions and commits made to any Eclipse Foundation project are covered by one of three distinct agreements – the Member Committer Agreement, the Individual Committer Agreement, or the Eclipse Contributor Agreement.

These agreements basically say that if you contribute to an Eclipse project, your contributions are being made under the license of the project. That license is usually the Eclipse Public License, but about 20% of our projects using additional or alternate licenses such as the Apache License, BSD, or MIT. It is important to note that the way things work at the Eclipse Foundation, the Foundation itself does not acquire any rights to the contributions. This is very different from other organizations like the FSF, OpenJDK, or the Apache Software Foundation. Eclipse uses a licensing model sometimes referred to as symmetrical inbound/outbound licensing, where contributors license their code directly to the users (recipients) of their contributions. Our approach requires us to ensure that all of our contribution agreements provide all necessary grants because we at the EF don’t have any rights to re-license contributions.

As most are aware, Eclipse is now about to start hosting specifications as open source projects.  This is very exciting for us, and we think it represents a new opportunity for creating innovative specifications using a vendor neutral process.  The first specification projects will be a part of the Jakarta EE initiative, but we expect other specification projects to follow shortly.

Everyone expected to re-sign one of these is encouraged to ensure they understand the details of the agreements and to seek their own legal advice. However, the change we have made is basically to ensure the copyrights in contributions to Eclipse projects may be used in specifications as well. (For the lawyers in the crowd, please note that these additional grants do not include patents.) We certainly expect that our committers and contributors are fine with this concept. In fact, I assume that most folks would have expected that this was already obvious when they contributed to an open source project. To that, all I can say is….ahhhh…the lawyers made us do it.

The new agreements are already posted, so they are in immediate effect for new contributors and committers. Since we need to overhaul our contribution agreements, we are also taking this opportunity to fix a few things. In particular, our committers will know that up until now they’ve been required to be covered by both a committer agreement and the ECA. We’re going to fix that, so if you sign an Individual Committer Agreement, or are covered by your employer’s Member Committer Agreement, you will no longer have to personally sign an ECA. We are also going to implementing electronic signatures for ICAs using HelloSign. So going forward there is going to be a little less paper involved in being a committer. Yay!

We’re sensitive that asking our contributors and committers to ‘update their paperwork’, especially if they’re not working on a specification, is – well, a pain in the backside.  But we’re hoping everyone will be supportive and understanding, and recognize that we take IP very seriously, and it’s one of the real value propositions of working with Eclipse.

Contributors who have an ECA will see them revoked over the coming months, and will be asked to re-sign the new one. We will be starting first with the contributors to the EE4J projects, since they are the ones who are most likely to have contributions flowing into Jakarta EE specifications.

Executing this change represents a massive effort for our team, as it literally means updating hundreds of committer agreements.  Our staff will be emailing individually each individual and member company needing to update their agreement with us, but we will be spread it over a period of the next few months.  So don’t be surprised if you don’t get an email for a while – we will get to everyone as soon as we can.

Stay tuned for emails on this subject that will be sent to our various mailing lists with more details.  If you have questions, feel free to reach out to us at license@eclipse.org and we’ll do our best to provide answers.

I thank our entire community in advance for accommodating this significant change.  We are excited about the Eclipse Foundation hosting an even more vibrant collection of projects, and believe hosting open source specification projects is a great step forward in our evolution!

Written by Mike Milinkovich

November 5, 2018 at 1:27 pm

Introducing the Jakarta EE Specification Process

I am very happy to announce that we are publishing a draft of the Eclipse Foundation Specification Process for community review and feedback. This specification process will be used by Jakarta EE as the new open specification process, replacing the JCP process previously used for Java EE. It is also expected that this new process will be of interest to other Eclipse working groups.

We are really looking forward to your feedback, which you can do via the Jakarta EE Community mailing list (preferred), or on the document comments.  The feedback provided will be used as input to finalizing a first version of the specification process and its adoption by Jakarta EE and other working groups at the Eclipse Foundation.  

As you are reviewing this draft specification process, please keep in mind the following key points about the approach that was taken by the Specification Committee.

  1. We want to design a specification process to replace the JCP. While there are many differences with the JCP, the key objective was to make the whole process as lightweight as possible.
  2. We want the specification process to be as close to open source development as possible. This is actually not a trivial exercise, as by its very nature drafting specifications is a somewhat different process.
  3. This is the Eclipse Spec Process, so we want to reuse the Eclipse Development Process wherever possible, and we want to ensure that the general flow and tone of the EDP is followed.
  4. We want to create a process that allows code-first development. Specifically, we want to enable a culture where experimentation can happen in open source and then have specifications be based on those experiences.
  5. We want the specifications that result from this process to be as high quality as possible. In particular, this means that we need to take care of the intellectual property flows, and to protect the community’s work from bad actors. This requirement manifests as two fundamentally important differences from the EDP:
    • Specification Committee approval is required for releases from Spec Projects, in addition to the normal PMC approval; and
    • We introduce the notion of “Participants” who are committers who represent specific member companies on a Spec Project. This is necessary to ensure that the IP contributions (particularly patents) from companies are properly captured by the process.

All of us at the Eclipse Foundation would like to recognize the tireless efforts of the members of the Specification Committee. A lot of hard work has gone into this document, and it’s very much appreciated. We are certain that Jakarta EE, and many other Eclipse technologies, benefit from the thoughtful efforts of this Committee.  In particular, we would like to thank the following Specification Committee members and alternates:

Fujitsu: Kenji Kazumura​, Mikel DeNicola
IBM: Dan Bandera​, Kevin Sutter
Oracle: Bill Shannon​, Ed Bratt​, Dmitry Kornilov
Payara: Steve Millidge​, Arjan Tijms
Red Hat: Scott Stark, Mark Little
Tomitribe: David Blevins​, Richard Monson-Haefel
PMC Representative: Ivar Grimstad
Elected Members: Alex Theedom, Werner Keil​

I also wish to recognize Tanja Obradovic and Wayne Beaton from the Eclipse Foundation team who have driven the process throughout – many thanks to you both!

Written by Mike Milinkovich

October 16, 2018 at 3:18 pm

Posted in Foundation, Jakarta EE, Open Source

Tagged with , ,

Jakarta EE Status – September 2018 Update

Migrating Java EE to the Eclipse Foundation and Jakarta EE is a process not an event. In the past couple of weeks however, several very important milestones have occurred that deserve to be recognized.

  • 100% of Glassfish and related Java EE reference implementation components from Oracle have now been contributed, and published to GitHub repositories of the EE4J organization. For those of us at the Eclipse Foundation, part of the reason why this is so huge is that to a large degree, we’ve completed our part. The repos (99) have been provisioned, the committers (162) have been given access, and the initial intellectual property reviews (404) have been done. From this point on, progress on the projects is now largely under the control of the projects themselves.

ee4j_status

  • Builds for the EE4J projects are now running on Eclipse Foundation infrastructure based on our Jenkins-based Common Build Infrastructure.
  • The Java EE TCKs have been contributed and are now available in open source. The importance of this cannot be understated, as my colleague Tanja Obradović points out in her blog.
  • The Eclipse Foundation has signed the Oracle Java EE TCK agreement, which is going to allow us to ship Eclipse Glassfish certified as Java EE 8 compatible. This has also required us to create a testing infrastructure at the Eclipse Foundation, and allowed the EE4J projects to begin testing against the Java EE 8 TCKs.
  • IBM, Oracle, Payara, Red Hat, and Tomitribe have all committed to three years of funding for Jakarta EE ranging from $25K to $300K per year. This funding will allow us to create a dedicated team and fund marketing activities for the Jakarta EE Working Group.

This migration has been an enormous effort, and we certainly have a ways to go yet. But it’s always fun to celebrate some victories along the way.

If you are interested in learning more about Jakarta EE and the future of cloud native Java, please join us at EclipseCon Europe in Germany in October. There is a wealth of Jakarta EE and MicroProfile content. In particular, I hope to see you at the talk that I am doing with Wayne Beaton about the new specification process. Oh, and there’s also this other conference happening in San Francisco at the same time….

 

Written by Mike Milinkovich

September 27, 2018 at 5:21 pm

Posted in Foundation, Open Source

K8s at the Edge – Some Context on the New Kubernetes IoT Working Group

It’s hard to believe that Google released Kubernetes as an open source project only three years ago. What began as the Borg cluster management platform to provide services like Gmail and YouTube at global scale is now the standard orchestration layer at the center of a massive industry shift to cloud native.

Enterprises are being catapulted into system resource engineering concepts that have been the bedrock of operations at web-scale leaders like Google and Netflix, and the open source stack underneath it all is evolving so fast it’s hard to keep up.  Kubernetes is attracting all sorts of exciting frameworks and services (see projects like Istio and Envoy), and no one could have anticipated it having this degree of momentum.

So what does any of this have to do with IoT?

The Eclipse Foundation is excited to be part of today’s announcement of a new Kubernetes IoT Edge Working Group – in conjunction with CNCF and supported by leading vendors like Bosch, Eurotech, InfluxData, Red Hat, Siemens, Vapor IO, and VMware.  

Eclipse Foundation has a lot on the go with IoT — 2.93 million lines of code (by last count), 42 different member organizations, 36 projects and 280 developers. The pace of innovation in IoT is very fast, and there are still a lot of great unsolved challenges at the IoT Edge.

Kubernetes was born in the datacenter. Its original promise was bridging datacenter and cloud.

But Kubernetes has a ton of potential for handling massive workloads on the edge — providing a common control plane across hybrid cloud and edge environments to simplify management and operations. Chick-fil-A, for example, is running Kubernetes at the edge in every restaurant.

Some of the great unsolved problems at the IoT edge — connectivity, manageability, scalability, reliability, security challenges, how to bring compute resources closer to edge devices for faster data processing and actions — are being solved as one-offs by the enterprises that are jumping into IoT. This new working group sees Kubernetes as having great potential as a foundational technology for extending hybrid environments to the IoT edge, and believes that broader industry collaboration on requirements definition around Kubernetes will accelerate broader adoption of IoT. Once again, it’s the rising tide theory of open source.

As Red Hat’s Dejan Bosanac (lead for the Kubernetes IoT Edge Working Group) says:

“IoT and edge applications have many distributed components that don’t usually sit together within the same datacenter infrastructure. There are messaging challenges, security has to be re-invented for every application and service, and there are integration and data locality issues with sidecar services. These are issues that shouldn’t have to be re-invented every time — they should be open source infrastructure with broad industry support. Red Hat and this working group see Kubernetes and other cloud-native projects in its orbit as having broad potential sitting between gateways, edge nodes and cloud platforms. Much like the LAMP stack was instrumental to the web-applications era, this group is focused on accelerating a Kubernetes stack for running cloud infrastructure and distributed components at the IoT edge.”

Some of the initial targets for the group in how it evolves Kubernetes for IoT edge applications:

  • Supporting Industrial IoT (IIoT) use cases scaling to millions of constrained devices a) connecting directly to Kubernetes-based cloud infrastructure (IP enabled devices), or b) connecting via IoT gateways (for non-IP enabled devices)
  • Via Edge nodes, bringing computing closer to data sources to support processing and acting on data sooner. Anticipated benefits include reduced latency, lower bandwidth, and improved reliability. Some example use cases:
    • Deploying data streaming applications to the Edge nodes in order to reduce traffic and save bandwidth between devices and the central cloud.
    • Deploying a serverless framework for using local functions that can be triggered as a response to certain events (without communication with the cloud)
  • Providing a common control plane across hybrid cloud and edge environments to simplify management and operations

The initial focus of the working group will be to flesh out IoT edge computing use cases and see how Kubernetes can be used (and to what extent). Among some of  the requirements identified so far:

  • For IIoT applications, the Kubernetes ingress layer must scale to millions of connections
  • That same ingestion layer must provide first-class support for IIoT messaging protocols such as MQTT (it is primarily HTTP TLS centric today)
  • Kubernetes must support multi-tenancy for environments where devices and gateways are shared

There have been a lot of industry watchers who have been wondering aloud how high the ceiling is for Kubernetes. Container orchestration seems like such a specific technology – how can this technology reach this type of ubiquity? Well one thing is clear – Kubernetes has jumped the datacenter and is now clearly in play in IoT ($1.2T market by 2022, according to IDC) as well. We are excited to be collaborating with the Kubernetes community to enable next generation of IoT applications.

To learn more, head over to the Kubernetes IoT Edge Working Group community page.

Have specific ideas for how you’d like to see Kubernetes evolve to meet your IoT use cases? Check out the this working document where you can get a sense of the progress so far, and the areas where you could possibly contribute your talents.

Written by Mike Milinkovich

September 26, 2018 at 7:00 am

Posted in Foundation, Open Source