ISSUE 122: Über Container Sprechen  

“Recently, the cost of RAM has continued to fall, and 64-bit processors and APIs enable in-memory data grids to be easily integrated with existing application and data layers, delivering in-memory speed plus massive scalability and high availability.”

In the race to make this weird, wild world of distributed, containerized applications compatible with the virtualized infrastructure upon which most enterprises depend, perhaps no project has made more progress than Kata Containers. The product of collaboration between the Hyper.sh project and Intel’s Clear Containers, Kata aims to pair individual containers with hypervisors, creating that direct link with the hardware that typifies the first-generation virtualization, and isolating host Linux kernels from one another.

Google’s recent gVisor project follows a similar path, creating a minimal Linux kernel for the container hosts that reduces the likelihood of exploit.

Some folks contend these architectures may render many of the more aggressive security systems being conceived for containerized environments unnecessary or redundant. But in a conversation for The New Stack Makers podcast, Aqua Security co-founder and CTO Amir Jerbi told us he believes that even the mode of process isolation that gVisor and Kata introduce would carry with it into practice some security challenges. Try orchestrating a microservices environment with isolated instances in a multitenant environment, he suggests, and see what happens.

Über Container Sprechen  

Though it wasn’t as large as vendor-centric conferences, ContainerDays, held in Hamburg this week, excelled in quality. Of course, the usual gifted and tireless speakers from the cloud-native conference circuit showed up — Cloud Native Computing Foundation (CNCF) Developer Advocate Ihor Dvoretskyi revealed that there are now more than 23,000 contributors to all CNCF projects, and Heptio head Craig McLuckie offered an interesting anecdote of how a mere three months passed after Google green-lighted Kubernetes as a project before K8s bits actually hit GitHub. Agile!

But the best talks revealed more about what was happening on this side of the pond, as us Yanks like to say. Maren Übelhör and Dr. Rebecca Dautel from the “anti-consultancy” firm Zoi GmbH spoke about how an organization could use Kubernetes to streamline an Apache Spark-based process of conducting data science tests. Sebastian Scheele, CTO and co-founder at Loodse, revealed a new project to make it easy for developers — not just operators — to use Kubernetes, called the Machine API. And Patrick Jahns, quality assurance lead for cloud storage provider ownCloud, offered a fascinating glimpse into that company’s move to DroneCI, a scalable containerized continuous integration system.

Also inspiring was the choice of venue. No dreary hotel conference center for these folks. Instead, the event was held at the Hafenmuseum, a museum dedicated to Hamburg’s rich heritage of being a major port for Germany. In the kickoff talk, the museum’s director Ursula Richenberger spoke about how in the last century shipping containers streamlined the shipping industry, eliminating the need for warehouses, such as the one this museum is now housed in, to shift goods from ships to trains or lorries. She quoted German politician Helmuth Kern as saying “The container is the package of globalization.”

As ContainerDays showed this week, history may very well repeat itself again.

Has Agile Programming Lost its Way?

Our weekend correspondent David Cassel caught up with one of the original creators of the Agile Manifesto, Ron Jeffries, who has been railing against “Faux Agile” — the misuse of principles he and others formulated for effective programming in the Agile manifesto. He argues that companies are misusing the concept as an attempt to push out more work from programmers.

IBM ‘BlockDrop’ Research Speeds Neural Networks

IBM Research has created a technology, called BlockDrop, that promises to speed convolutional neural network operations without any loss of fidelity. BlockDrop works by looking for layers in deep networks that aren’t necessary to compute to achieve the desired level of accuracy, and then dropping those layers on the fly, allowing the system to allocate resources in a more efficient manner. This could further excel the use of neural nets, particularly in places with limited computing capability.

Deploy a Multicluster Ingress on Google Kubernetes Engine

TNS analyst Janakiram MSV has been working on an ambitious series on Federated Kubernetes, the ability to run Kubernetes across different clusters. In this series he walks through the steps of setting up an ingress that can route the traffic to the nearest cluster based on a user’s location.