The New Stack Podcast

For this episode of The New Stack Makers, TC Currie is joined by Dr. Eric Brewer, Google Fellow and VP of Infrastructure at Google where he’s the architect of Cloud Native systems at massive scale.  He’s also been a Computer Science Professor at UC Berkeley since 1994.   TC caught up with Dr. Brewer at the recent Service Mesh Days in San Francisco.

“Istio at its core decouples developers from operations,” Brewer said. That decoupling lets both sides go faster and with more confidence.

Traditionally, when developers are writing a service, they worry a lot about the API, what are the methods, how does it work? But when you’re deploying microservices, then you start need to think about other questions.

What are the policies that are calling this service?  Does it have quota? Does it have a denial of service?  How does it get authenticated?  How is it secured?  All of these questions that are not about what the API does, but are operations pieces.

GitLab has been very busy recently expanding on its leading software repository manager for DevOps to underpin its customers' continuous integration/continuous delivery (CI/CD) efforts.

GitLab is also a very transparent company and actively communicates lessons it has learned along the way firsthand to better support its customers. As a case in point, Priyanka Sharma, director of technical evangelism at GitLab who serves on the board of the Cloud Native Computing Foundation (CNCF), recently surprised many by describing in a recent The New Stack blog post how GitLab began its CD journey without first completely shifting its underlying IT infrastructure to Kubernetes first.

“We’re in the process of moving to Kubernetes, and we need to deploy even more frequently than we have in the past as we increase the velocity of feature development,” Sharma wrote. “But instead of modernizing completely with Kubernetes and then starting CD, we have opted to push our existing CI/CD system to the limit by using our preexisting legacy tools — and a lot of smarts.”

In this episode of The New Stack Makers podcast hosted by Alex Williams, founder and editor in chief of The New Stack, Sharma further discussed how GitLab has sought to not necessarily opt for necessarily the latest-and-greatest tools sets in its move to CD. Instead, the focus has been on finding the most direct path while maintaining the investments in and support of GitLab’s legacy systems.

“We're now in a place where we will start slowly moving to canaries because now we're maxed out what we could do with this legacy system,” Sharma said.

This week we spoke with Chip Childers, technology chief of staff at the Cloud Foundry Foundation, about the technical and architectural decisions that help define the modern enterprise. Cloud Foundry Foundation is the independent organization that oversees the development and adoption of the Cloud Foundry platform, and it’s a sponsor of The New Stack.

More than half of the Fortune 500 uses Cloud Foundry, so the Cloud Foundry Foundation depends on the enterprise using, contributing to, and advancing its open source projects. Childers works closely with developers and CIOs at all of the foundation’s member companies, and so he has a deep understanding of what it is that a modern enterprise needs in order to succeed in today’s rapidly changing technology landscape. In a recent post for The New Stack, he wrote that  “Modern enterprises ... think critically about what they should build themselves and what they should source from somewhere else.”

We've written some about accessibility in terms of building apps and websites — although there's always more to learn and do — but what about accessibility and inclusion in the developer world? How can you best serve your full customer base when your customers are developers? How is designing for inclusion an essential part of the developer experience? These are the questions Pivotal Labs' Senior Designer Raquel Breternitz is trying to answer in this episode of The New Stack Makers.

Why is inclusion so important to Breternitz?

"We don't have to just make sure that someone can get there, and can understand it when they do, but we also have to find the ways in which folks are excluded and break those barriers down as well," she told The New Stack in a follow-up conversation.

Splice Machine brings SQL Back, said Monte Zweben, CEO at Splice Machine, but with a spectacular twist.

It’s common for data infrastructure to reside on three separate platforms:  transactional databases, data warehouses, and the recently-added machine learning (ML) environment.

Most artificial intelligence (AI) applications are built on historical data that has been loaded in to ML environments for data mining.  So you’re looking in the rearview mirror to figure out what has already happened, he said.  And that requires computational engines.

Then, since you want to perform actions to make decisions in the moment, you need an operational data store, like traditional database management systems or a NoSQL data store, but people use these to make decisions in the application in the moment.

Lastly is new ML algorithms, data science workbenches and modeling workflows.

This week, we talk with Charity Majors, chief technology officer of Honeycomb.io and a pioneer in observability about the three-year anniversary of that movement which has significantly changed how applications are monitored and maintained in the cloud native era.

Majors contributed a post that appeared on The New Stack this week in which she takes a closer look at why the observability movement formed and why other approaches and methods fall short. She explained why engineering teams must continue to focus on this approach in order to better understand, debug, and maintain large-scale distributed systems.

"When we blew up the monolith into many services, we lost the ability to step through our code with a debugger: it now hops the network. Our tools are still coming to grips with this seismic shift," Majors writes.

It’s just easier — a greenfield startup founder picks and chooses the best technology platforms and software development tools that best serve the needs of the business model. For organizations with years of legacy software in place, the digital journey is often harder, yet inherently necessary to survive in today’s DevOps-centric world.

In this episode of The New Stack Makers podcast hosted by Alex Williams, founder and editor in chief of The New Stack, ahead of the upcoming TC Sessions: Enterprise event, how organizations attempt to keep up with the disruption amid an explosion of new and powerful open source tools served as a central theme.

On hand to discuss what is at stake were:

Ron Miller, enterprise reporter at TechCrunch.

Yvonne Wassenaar, CEO of Puppet.

Shawn Wormke, senior director of Aspen Mesh.

What does Justin Bieber have to do with our relationships to, well, anything? If nothing else, his staggering 105.8 million followers on Twitter (at the time of this conversation) is a real challenge for data scientists. If you wanted to know how The Bieb’s assemblage relates to him and one another, you could never achieve it with an entity-first relationship model — you know, columns, rows and spreadsheets, or, as Wikipedia calls it, “a simple relational database implementation, each row of a table represents one instance of an entity type, and each field in a table represents an attribute type.” If you really want to begin to understand the reasoning behind The Beib’s broad fanbase, you’d need to organize data relationship-first.

This is just one of the real-world examples Dr. Denise Gosnell offered regarding the importance of graph databases and the tech behind them that is shifting how we think about and organize data. The head of global Graph practice at DataStax and one of the world’s foremost researchers on graph theory, graph algorithms, graph databases, and applications of graph data across all industry verticals offered more business use cases as well, including the interrelatedness of your primary and secondary LinkedIn connections and the growing popularity of smart meters.

Welcome to The New Stack Context, a podcast where we discuss the latest news and views from the cloud native community. This week, we spoke Luca Mezzalira, chief architect at sports video streaming site DAZN (Da-Zone), about the idea and evolution of microfrontends, or the equivalent of a microservices approach to scaling frontend development.

Mezzalira has written and presented talks about his journey to design the company’s scale-out frontend streaming application, one that he calls one of the most challenging architectures he’s ever created. DAZN was dealing with a lot of the same issues that companies with large, monolithic backends face as they tried to quickly, and massively scale. We’ll hear about how his team developed the DAZN microfrontends. If content is king, then distribution is the queen, he reminded us in this lively interview.

Then later in the show, we’ll discuss the recent data breach at Capital One due to a “firewall misconfiguration” that allowed a hacker to access data on over 100 million customers in the company’s AWS S3 buckets. As regular readers know, The New Stack has had dozens of posts (most recently) about the dangers of cloud security over the past few years. And they all have the same basic underlying message: even if you use third-party cloud services, security is still your responsibility.

“There’s so much passion for using AI in ways that are beneficial,” Anna Bethke, Head of AI4 Social Good at Intel’s AI Products Group.  And she wanted to make it easier for others to get involved.

Bethke’s team partners with companies and social justice organizations to give them access to Intel’s knowledge of deep learning skills and techniques, making it easier for them to mine social media and other data to filter out the noise.

One of her first partnerships is with National Center for Missing and Exploited Children (NCMEC). Their mission is to find children who are being trafficked or in danger of being kidnapped and alert the local law enforcement so they can take action. They get hundreds of thousands of alerts a day from social media like Facebook and Twitter, which was completely unmanageable.

There are three points of information NCMEC needs.  First, identify and prioritize which posts represent the most danger to the child.  Second, identify where the post is originating, which is often complicated by the use of multiple IP addresses and IP address obscuring.  Once the location is found, the legal jurisdiction needs to be identified so the information can be sent to the proper authorities (e.g., police or sheriff department).  But how do you find that information in the flood of data?