The New Stack Analysts

Serverless compute sounds like heaven, but without the right mindset it can be hell. Between the constraints placed on serverless environments by cloud providers, the potential for proliferation of applications and variations of versions, and the general mind shift required to build for serverless, the shift to cloud functions isn't something that just happens overnight.

"The interesting thing about serverless platforms is they are a constrained execution environment, which is a good thing. It means that the cloud vendor can provide capabilities for customers at a fraction of the cost of what they were if they were fully customizable... The problem is, as these platforms are still maturing, they are necessarily constrained in what kind of execution they will allow. For instance, most of them don't let you run for more than five or ten minutes... The challenge is, what if you can't run in that uniform environment? This is where containers come into play," said  Donna Malayeri, Product Manager at Pulumi.

"OK, we're going to write some functions, but how do we deal with our data? We have 14 petabytes of data and we can't just whimsically move it somewhere. So, introducing more environment control from that perspective is kind of a big deal to me. I keep running into these scenarios, and a lot of the times it's still a big question mark. What are we going to do? We started to build on this new serverless architecture, but we still have huge data problems," said Adron Hall, of Thrashing Code.

Serverless, containers? The debate doesn't matter for customers coming to ioPipe, says Erica Windisch, founder and CTO of ioPipe. Their customers have already decided to adopt or even further adopt serverless architectures. They are not talking about containers or Kuberbetes. Her comments speak to a deeper discussion about the role of container and serverless architectures that TNS Analysts Guest Host Adron Hall addresses in discussing the larger picture about the decisions any company is trying to make now.

"I have been hearing more about trying to run Open FaaS in Kubernetes, which is an interesting take on the whole situation," said Hall.

Full cloud services offer it all but there are also applications running on bare metal, others in VMs, across a heterogeneous infrastructure and hybrid/CDN-style hosts. It's a deeper discussion that will help me at least think of guide posts as we talk to more people who are making decisions about serverless or not. It may all come down to what Windisch describes as "application operations."

To do cloud-native computing, you need to identify all your workloads, and, more importantly, they need the ability to identify each other, so they can work together in automated chains. To aid in this task, the Cloud Native Computing Foundation has adopted the open source SPIFFE specification, and its associated SPIRE runtime. SPIFFE provides a standard for securely identifying software components in heterogeneous IT systems and SPIRE is the engine that can make it happen (and, in this setup,  CNCF's Open Policy Agent [OPA] can enforce the authorization duties).

If you feel all this is a bit much to take in, then you are not alone.  For our latest "pancakes and podcast" edition of the The New Stack Analysts — recorded live at the Kubecon + CloudNativeCon Europe 2018 on May 3 — we focused our panel discussion on SPIFFE, and the room was filled with those curious about this topic (and/or hungry for delicious pancakes).

Watch on YouTube: https://youtu.be/lO7CZZVGRz4

For this latest edition of The New Stack Analysts, we took our pancakes and our podcast equipment to Boston, for the Cloud Foundry Summit in Boston, for a wide ranging discussion on Cloud Foundry, cloud-native computing and Kubernetes.

Hosted by TNS founder Alex Williams, with TNS managing editor Joab Jackson, our panel consisted of:

Frederic Lardinois, reporter for TechCrunch.

Abby Kerns, executive director for Cloud Foundry Foundation.

Chen Goldberg, engineering director for Google.

Jennifer Kotzen, senior product manager for SUSE.

Watch on YouTube: https://youtu.be/ekBh7_9ZVYA

How venture capitalists and technologists view the cloud-native technology space is the subject matter for this episode of The New Stack Analysts.

It’s getting to the point that service level agreements pretty much don’t allow access to source code unless it’s open source. To Lightstep Co-Founder Ben Sigelman that pretty much means that open-source code is now part of business and it’s a matter of developing the code into stadard techologies that may be used by all.

Sigleman was joined by RedPoint Venture Capital’s Astasia Myers and Scott Raney who say there is no way to attack the infrastructure layer market unless you have a core belief in the role of open source. RedPoint is a co-author of the cloud native landscape run in conjunction with the Cloud Native Computing Foundation.

On today's episode of The New Stack Analysts, TNS Founder Alex Williams, TNS Correspondent TC Curie, and  Janakiram MSV, Principal Analyst at Janakiram & Associates were joined by Heptio Co-Founder and CTO and Kubernetes co-founder Joe Beda, alongside Sebastien Goasguen, Kubernetes Tech Lead at Bitnami. The discussion this week centered around the many abstractions available to developers working with Kubernetes, and how these impact developer teams both large and small.

“What I’m seeing is there is this full effort to bring in another abstraction layer on top of Kubernetes to encourage users, beginners, and even large enterprise IT teams to target Kubernetes without understanding the nuts and bolts of Kubernetes certificates," said MSV.

The concepts of craft and how it applies to software development is quite relevant but at times contradictory.  Technologists aspire to make great tools and services that reflect their attention and passion for making things but paradoxically we also face a culture with values based upon speed and the mass consumption of resources.

How craft relate to the way we work and how is it relevant in a culture that puts such a premium on speed and consumption is what we discuss in this episode of The New Stack Analysts with RedMonk Co-Founder James Governor and Charity Majors, Honeycomb.io co-founder and CEO. It's also a core theme for Monkigras, the conference in London that Governor is hosting this week and where Majors is one of the speakers.

It's this idea of sustainable craft that Governor is building as a core theme of Monkigras. The conference will put an emphasis on "topics such as apprenticeship, community management, writing sustainable documentation, software archaeology, sustainable open source, hop farming, and typefaces."

In the second instalment of our podcasts on Values and how they affect workplace culture, Alex Williams and TC Currie from The New Stack are joined by Sam Ramji, VP Product Management at Google, Daniel Lopez Ridruejo, CEO Bitnami; Chris Brandon, CEO, StorageOS and Dave McCrory, VP of Software Engineering at GE Digital. Their companies range in size from seventeen employees to many thousands, but the need for consistent values across the company, no matter its size, is universal. “What we’ve found at The New Stack,” said Williams, “is it all comes down to trust, respect, and integrity.”

Google’s mission, said Ramji, “is to organize the world’s information and make it universally accessible and universally useful.” This core value is overlaid at the engineering level with the focus of “How are we doing this at the next order of magnitude?” As a company, they are working on how to provide engineers with a level of empathy for individual engineers.

Listen to part one of our miniseries on Values at:

https://soundcloud.com/thenewstackanalysts/154-how-do-values-affect-software-companies

In this livestream from KubeCon + CloudNativeCon 2017, The New Stack Analysts will take a deep dive into Univa’s Universal Resource Broker (URB). We learned how this open source adapter allows Kubernetes and Mesos workloads to run on the same infrastructure, where TNS Founder Alex Williams was joined by Univa Corporation CTO and Business Developer Fritz Ferstl, and Thrashing Code Software Architect Adron Hall.

"I know a lot of groups want to containerise [their applications]. Bring over the modularisation and isolation without worrying about the 30 year old hardware that it runs on," said Hall.

Ferstl added that people running Kubernetes rather than Mesos may look at it and say, "There is something useful there, I can work with that, rather than try to force code into Kubernetes, I can take something like the URB that just runs it."

Watch on YouTube: https://youtu.be/sIkzQC9DOSc

In this end-of-the-year discussion about missions, values and goals, Alex Williams, TNS Founder was joined by Ashley Williams, Services and Ops Engineer at npm, Inc. who has been active in the Node.js and Rust communities, Charity Majors, CEO of Honeycomb, a company that provides “observability for a distributed world,” and TC Currie, TNS San Francisco Correspondent.

Topics include hacker culture, preferred perks, what gets lost when pushing employees, what is gained and more.