More than 12,000 attendees gathered this week in San Diego to discuss all things containers, Kubernetes and cloud-native at KubeCon.

Kubernetes, the container orchestration tool, turned five this year, and the technology appears to be reaching a maturity phase where it accelerates beyond early adopters to reach a more mainstream group of larger business users.

That’s not to say that there isn’t plenty of work to be done, or that most enterprise companies have completely bought in, but it’s clearly reached a point where containerization is on the table. If you think about it, the whole cloud-native ethos makes sense for the current state of computing and how large companies tend to operate.

If this week’s conference showed us anything, it’s an acknowledgment that it’s a multi-cloud, hybrid world. That means most companies are working with multiple public cloud vendors, while managing a hybrid environment that includes those vendors — as well as existing legacy tools that are probably still on-premises — and they want a single way to manage all of this.

The promise of Kubernetes and cloud-native technologies, in general, is that it gives these companies a way to thread this particular needle, or at least that’s the theory.

Kubernetes to the rescue

Photo: Ron Miller/TechCrunch

If you were to look at the Kubernetes hype cycle, we are probably right about at the peak where many think Kubernetes can solve every computing problem they might have. That’s probably asking too much, but cloud-native approaches have a lot of promise.

Craig McLuckie, VP of R&D for cloud-native apps at VMware, was one of the original developers of Kubernetes at Google in 2014. VMware thought enough of the importance of cloud-native technologies that it bought his former company, Heptio, for $550 million last year.

As we head into this phase of pushing Kubernetes and related tech into larger companies, McLuckie acknowledges it creates a set of new challenges. “We are at this crossing the chasm moment where you look at the way the world is — and you look at the opportunity of what the world might become — and a big part of what motivated me to join VMware is that it’s successfully proven its ability to help enterprise organizations navigate their way through these disruptive changes,” McLuckie told TechCrunch.

He says that Kubernetes does actually solve this fundamental management problem companies face in this multi-cloud, hybrid world. “At the end of the day, Kubernetes is an abstraction. It’s just a way of organizing your infrastructure and making it accessible to the people that need to consume it.

“And I think it’s a fundamentally better abstraction than we have access to today. It has some very nice properties. It is pretty consistent in every environment that you might want to operate, so it really makes your on-prem software feel like it’s operating in the public cloud,” he explained.

Simplifying a complex world

One of the reasons Kubernetes and cloud-native technologies are gaining in popularity is because the technology allows companies to think about hardware differently. There is a big difference between virtual machines and containers, says Joe Fernandes, VP of product for Red Hat cloud platform.

“Sometimes people conflate containers as another form of virtualization, but with virtualization, you’re virtualizing hardware, and the virtual machines that you’re creating are like an actual machine with its own operating system. With containers, you’re virtualizing the process,” he said.

He said that this means it’s not coupled with the hardware. The only thing it needs to worry about is making sure it can run Linux, and Linux runs everywhere, which explains how containers make it easier to manage across different types of infrastructure. “It’s more efficient, more affordable, and ultimately, cloud-native allows folks to drive more automation,” he said.

Bringing it into the enterprise

Photo: Ron Miller/TechCrunch

It’s one thing to convince early adopters to change the way they work, but as this technology enters the mainstream. Gabe Monroy, partner program manager at Microsoft says to carry this technology to the next level, we have to change the way we talk about it.

Software will eat the world, as the saying goes, but in doing so, some developers are likely to get a little indigestion. That is to say, building products requires working with disparate and distributed teams, and while developers may have an ever-growing array of algorithms, APIs and technology at their disposal to do this, ironically the platforms to track it all haven’t evolved with the times. Now three developers have taken their own experience of that disconnect to create a new kind of platform, Linear, which they believe addresses the needs of software developers better by being faster and more intuitive. It’s bug tracking you actually want to use.

Today, Linear is announcing a seed round of $4.2 million led by Sequoia, with participation also from Index Ventures and a number of investors, startup founders and others that will also advise Linear as it grows. They include Dylan Field (Founder and CEO, Figma), Emily Choi (COO, Coinbase), Charlie Cheever (Co-Founder of Expo & Quora), Gustaf Alströmer (Partner, Y Combinator), Tikhon Berstram (Co-Founder, Parse), Larry Gadea (CEO, Envoy), Jude Gomila (CEO, Golden), James Smith (CEO, Bugsnag), Fred Stevens-Smith (CEO, Rainforest), Bobby Goodlatte, Marc McGabe, Julia DeWahl and others.

Cofounders Karri Saarinen, Tuomas Artman, and Jori Lallo — all Finnish but now based in the Bay Area — know something first-hand about software development and the trials and tribulations of working with disparate and distributed teams. Saarinen was previously the principal designer of Airbnb, as well as the first designer of Coinbase; Artman had been staff engineer and architect at Uber; and Lallo also had been at Coinbase as a senior engineer building its API and front end.

“When we worked at many startups and growth companies we felt that the tools weren’t matching the way we’re thinking or operating,” Saarinen said in an email interview. “It also seemed that no-one had took a fresh look at this as a design problem. We believe there is a much better, modern workflow waiting to be discovered. We believe creators should focus on the work they create, not tracking or reporting what they are doing. Managers should spend their time prioritizing and giving direction, not bugging their teams for updates. Running the process shouldn’t sap your team’s energy and come in the way of creating.”

Linear cofounders (from left): KarriSaarinen, Jori Lallo, and Tuomas Artma

All of that translates to, first and foremost, speed and a platform whose main purpose is to help you work faster. “While some say speed is not really a feature, we believe it’s the core foundation for tools you use daily,” Saarinen noted.

A ⌘K command calls up a menu of shortcuts to edit an issue’s status, assign a task, and more so that everything can be handled with keyboard shortcuts. Pages load quickly and synchronise in real time (and search updates alongside that). Users can work offline if they need to. And of course there is also a dark mode for night owls.

The platform is still very much in its early stages. It currently has three integrations based on some of the most common tools used by developers — GitHub (where you can link Pull Requests and close Linear issues on merge), Figma designs (where you can get image previews and embeds of Figma designs), and Slack (you can create issues from Slack and then get notifications on updates). There are plans to add more over time.

Indeed, speed is not the only aim with Linear. Saarinen also said another area they hope to address is general work practices, with a take that seems to echo a turn away from time spent on manual management and more focus on automating that process.

“Right now at many companies you have to manually move things around, schedule sprints, and all kinds of other minor things,” he said. “We think that next generation tools should have built in automated workflows that help teams and companies operate much more effectively. Teams shouldn’t spend a third or more of their time a week just for running the process.”

The last objective Linear is hoping to tackle is one that we’re often sorely lacking in the wider world, too: context.

“Companies are setting their high-level goals, roadmaps and teams work on projects,” he said. “Often leadership doesn’t have good visibility into what is actually happening and how projects are tracking. Teams and contributors don’t always have the context or understanding of why they are working on the things, since you cannot follow the chain from your task to the company goal. We think that there are ways to build Linear to be a real-time picture of what is happening in the company when it comes to building products, and give the necessary context to everyone.”

Linear is a late entrant in a world filled with collaboration apps, and specifically workflow and collaboration apps targeting the developer community. These include not just Slack and GitHub, but Atlassian’s Trello and Jira, as well as Asana, Basecamp and many more.

Saarinen would not be drawn out on which of these (or others) that it sees as direct competition, noting that none are addressing developer issues of speed, ease of use and context as well as Linear is.

“There are many tools in the market and many companies are talking about making ‘work better,’” he said. “And while there are many issue tracking and project management tools, they are not supporting the workflow of the individual and team. A lot of the value these tools sell is around tracking work that happens, not actually helping people to be more effective. Since our focus is on the individual contributor and intelligent integration with their workflow, we can support them better and as a side effect makes the information in the system more up to date.”

Stephanie Zhan, the partner at Sequoia whose speciality is seed and Series A investments and who has led this round, said that Linear first came on her radar when it first launched its private beta (it’s still in private beta and has been running a waitlist to bring on new users. In that time it’s picked up hundreds of companies, including Pitch, Render, Albert, Curology, Spoke, Compound and YC startups including Middesk, Catch and Visly). The company had also been flagged by one of Sequoia’s Scouts, who invested earlier this year

Although Linear is based out of San Francisco, it’s interesting that the three founders’ roots are in Finland (with Saarinen in Helsinki this week to speak at the Slush event), and brings up an emerging trend of Silicon Valley VCs looking at founders from further afield than just their own back yard.

“The interesting thing about Linear is that as they’re building a software company around the future of work, they’re also building a remote and distributed team themselves,” Zahn said. The company currently has only four employees.

In that vein, we (and others, it seems) had heard that Sequoia — which today invests in several Europe-based startups, including Tessian, Graphcore, Klarna, Tourlane, Evervault  and CEGX — has been considering establishing a more permanent presence in this part of the world, specifically in London.

Sources familiar with the firm, however, tell us that while it has been sounding out VCs at other firms, saying a London office is on the horizon might be premature, as there are as yet no plans to set up shop here. However, with more companies and European founders entering its portfolio, and as more conversations with VCs turn into decisions to make the leap to help Sequoia source more startups, we could see this strategy turning around quickly.

At its Cloud Next event in London, Google today announced a number of product updates around its managed Anthos platform, as well as Apigee and its Cloud Code tools for building modern applications that can then be deployed to Google Cloud or any Kubernetes cluster.

Anthos is one of the most important recent launches for Google, as it expands the company’s reach outside of Google Cloud and into its customers’ data centers and, increasingly, edge deployments. At today’s event, the company announced that it is taking Anthos Migrate out of beta and into general availability. The overall idea behind Migrate is that it allows enterprises to take their existing, VM-based workloads and convert them into containers. Those machines could come from on-prem environments, AWS, Azure or Google’s Compute Engine, and — once converted — can then run in Anthos GKE, the Kubernetes service that’s part of the platform.

“That really helps customers think about a leapfrog strategy, where they can maintain the existing VMs but benefit from the operational model of Kubernetes,” Google VP of product management Jennifer Lin told me. “So even though you may not get all of the benefits of a cloud-native container day one, what you do get is consistency in the operational paradigm.”

As for Anthos itself, Lin tells me that Google is seeing some good momentum. The company is highlighting a number of customers at today’s event, including Germany’s Kaeser Kompressoren and Turkey’s Denizbank.

Lin noted that a lot of financial institutions are interested in Anthos. “A lot of the need to do data-driven applications, that’s where Kubernetes has really hit that sweet spot because now you have a number of distributed datasets and you need to put a web or mobile front end on [them],” she explained. “You can’t do it as a monolithic app, you really do need to tap into a number of datasets — you need to do real-time analytics and then present it through a web or mobile front end. This really is a sweet spot for us.”

Also new today is the general availability of Cloud Code, Google’s set of extensions for IDEs like Visual Studio Code and IntelliJ that helps developers build, deploy and debug their cloud-native applications more quickly. The idea, here, of course, is to remove friction from building containers and deploying them to Kubernetes.

In addition, Apigee hybrid is now also generally available. This tool makes it easier for developers and operators to manage their APIs across hybrid and multi-cloud environments, a challenge that is becoming increasingly common for enterprises. This makes it easier to deploy Apigee’s API runtimes in hybrid environments and still get the benefits of Apigees monitoring and analytics tools in the cloud. Apigee hybrid, of course, can also be deployed to Anthos.

Deep learning is all the rage these days in enterprise circles, and it isn’t hard to understand why. Whether it is optimizing ad spend, finding new drugs to cure cancer, or just offering better, more intelligent products to customers, machine learning — and particularly deep learning models — have the potential to massively improve a range of products and applications.

The key word though is ‘potential.’ While we have heard oodles of words sprayed across enterprise conferences the last few years about deep learning, there remain huge roadblocks to making these techniques widely available. Deep learning models are highly networked, with dense graphs of nodes that don’t “fit” well with the traditional ways computers process information. Plus, holding all of the information required for a deep learning model can take petabytes of storage and racks upon racks of processors in order to be usable.

There are lots of approaches underway right now to solve this next-generation compute problem, and Cerebras has to be among the most interesting.

As we talked about in August with the announcement of the company’s “Wafer Scale Engine” — the world’s largest silicon chip according to the company — Cerebras’ theory is that the way forward for deep learning is to essentially just get the entire machine learning model to fit on one massive chip. And so the company aimed to go big — really big.

Today, the company announced the launch of its end-user compute product, the Cerebras CS-1, and also announced its first customer of Argonne National Laboratory.

The CS-1 is a “complete solution” product designed to be added to a data center to handle AI workflows. It includes the Wafer Scale Engine (or WSE, i.e. the actual processing core) plus all the cooling, networking, storage, and other equipment required to operate and integrate the processor into the data center. It’s 26.25 inches tall (15 rack units), and includes 400,000 processing cores, 18 gigabytes of on-chip memory, 9 petabytes per second of on-die memory bandwidth, 12 gigabit ethernet connections to move data in and out of the CS-1 system, and sucks just 20 kilowatts of power.

A cross-section look at the CS-1. Photo via Cerebras

Cerebras claims that the CS-1 delivers the performance of more than 1,000 leading GPUs combined — a claim that TechCrunch hasn’t verified, although we are intently waiting for industry-standard benchmarks in the coming months when testers get their hands on these units.

In addition to the hardware itself, Cerebras also announced the release of a comprehensive software platform that allows developers to use popular ML libraries like TensorFlow and PyTorch to integrate their AI workflows with the CS-1 system.

In designing the system, CEO and co-founder Andrew Feldman said that “We’ve talked to more than 100 customers over the past year and a bit,“ in order to determine the needs for a new AI system and the software layer that should go on top of it. “What we’ve learned over the years is that you want to meet the software community where they are rather than asking them to move to you.”

I asked Feldman why the company was rebuilding so much of the hardware to power their system, rather than using already existing components. “If you were to build a Ferrari engine and put it in a Toyota, you cannot make a race car,” Feldman analogized. “Putting fast chips in Dell or [other] servers does not make fast compute. What it does is it moves the bottleneck.” Feldman explained that the CS-1 was meant to take the underlying WSE chip and give it the infrastructure required to allow it to perform to its full capability.

A diagram of the Cerebras CS-1 cooling system. Photo via Cerebras.

That infrastructure includes a high-performance water cooling system to keep this massive chip and platform operating at the right temperatures. I asked Feldman why Cerebras chose water, given that water cooling has traditionally been complicated in the data center. He said, “We looked at other technologies — freon. We looked at immersive solutions, we looked at phase-change solutions. And what we found was that water is extraordinary at moving heat.”

A side view of the CS-1 with its water and air cooling systems visible. Photo via Cerebras.

Why then make such a massive chip, which as we discussed back in August, has huge engineering requirements to operate compared to smaller chips that have better yield from wafers. Feldman said that “ it massively reduces communication time by using locality.”

In computer science, locality is placing data and compute in the right places within, let’s say a cloud, that minimizes delays and processing friction. By having a chip that can theoretically host an entire ML model on it, there’s no need for data to flow through multiple storage clusters or ethernet cables — everything that the chip needs to work with is available almost immediately.

According to a statement from Cerebras and Argonne National Laboratory, Cerebras is helping to power research in “cancer, traumatic brain injury and many other areas important to society today” at the lab. Feldman said that “It was very satisfying that right away customers were using this for things that are important and not for 17-year-old girls to find each other on Instagram or some shit like that.”

(Of course, one hopes that cancer research pays as well as influencer marketing when it comes to the value of deep learning models).

Cerebras itself has grown rapidly, reaching 181 engineers today according to the company. Feldman says that the company is hands down on customer sales and additional product development.

It has certainly been a busy time for startups in the next-generation artificial intelligence workflow space. Graphcore just announced this weekend that it was being installed in Microsoft’s Azure cloud, while I covered the funding of NUVIA, a startup led by the former lead chip designers from Apple who hope to apply their mobile backgrounds to solve the extreme power requirements these AI chips force on data centers.

Expect ever more announcements and activity in this space as deep learning continues to find new adherents in the enterprise.