New tool uses AI to roll back problematic continuous delivery builds automatically

As companies shift to CI/CD (continuous integration/continuous delivery), they face a problem around monitoring and fixing problems in builds that have been deployed. How do you deal with an issue after moving onto the next delivery milestone? Harness, the startup launched last year by AppDynamics founder Jyoti Bansal, wants to fix that with a new […]

As companies shift to CI/CD (continuous integration/continuous delivery), they face a problem around monitoring and fixing problems in builds that have been deployed. How do you deal with an issue after moving onto the next delivery milestone? Harness, the startup launched last year by AppDynamics founder Jyoti Bansal, wants to fix that with a new tool called 24×7 Service Guard.

The new tool is designed to help companies working with a continuous delivery process by monitoring all of the builds, regardless of when they were launched. What’s more, the company claims that using AI and machine learning, it can dial back a problematic build to one that worked in an automated fashion, freeing developers and operations to keep working without worry.

The company launched last year with a tool called Continuous Verification to verify that a continuous delivery build got deployed. With today’s announcement, Bansal says the company is taking this to another level to help understand what happens after you deploy.

The tool watches every build, even days after deployment, taking advantage of data from tools like AppDynamics, New Relic, Elastic and Splunk, then using AI and machine learning to identify problems and bring them back to a working state without human intervention. What’s more, your team can get a unified view of performance and the quality of every build across all of your monitoring and logging tools.

“People are doing Continuous Delivery and struggling with it. There are also using these AI Ops kinds of products, which are watching things in production, and trying to figure out what’s wrong. What we are doing is we’re bringing the two together and ensuring nothing goes wrong,” Bansal explained.

24×7 Service Guard Console. Screenshot: Harness

He says that he brought this product to market because he saw enterprise companies struggling with CI/CD. He said the early messaging that you should move fast and break things really doesn’t work in enterprise settings. They need tooling that ensures that critical applications will keep running even with continuous builds (however you define that). “How do you enable developers so that they can move fast and make sure the business doesn’t get impacted. I feel that industry was underserved by this [earlier] message,” he said.

While it’s hard for any product to absolutely guarantee up-time, this one is providing tooling for companies who see the value of CI/CD, but are looking for a way to keep their applications up and running, so they aren’t constantly on this deploy/repair treadmill. If it works as described, it could help advance CI/CD, especially for large companies who need to learn to move faster and want assurances that when things break, they can be fixed in an automated fashion.

Drone.io, Packet team on free continuous delivery service for open source developers

Drone.io, makers of the open source Drone continuous integration/continuous delivery tool (CI/CD), announced Drone Cloud today, a new CI/CD cloud service that it’s making available for free to open source projects. The company is teaming with Packet, which is offering to run the service for free on its servers. Drone.io co-founder Brad Rydzewski says his […]

Drone.io, makers of the open source Drone continuous integration/continuous delivery tool (CI/CD), announced Drone Cloud today, a new CI/CD cloud service that it’s making available for free to open source projects. The company is teaming with Packet, which is offering to run the service for free on its servers.

Drone.io co-founder Brad Rydzewski says his company is “a container-native continuous delivery platform, and its goal is to help automate the developer workflow from testing to release.” Continuous delivery is an approach built on cloud-native, the idea that you can manage cloud and on prem with single set of management tools. From a developer standpoint, it relies on containers as a way to continuously deliver application updates as changes happen.

As part of that approach, the newly announced Drone Cloud provides a publicly hosted CI/CD cloud offering. “It’s free for the open source community. So it’s an open source only offering. There’s no paid plan, and it’s only available to public Github repositories,” Rydzewski explained.

Rydzewski says the service was born out of a need for his own project. He found it hard to find publicly-hosted solutions that offered a way to test a variety of operating systems and chip architectures. “It’s really something we needed for ourselves. The Drone community wanted to run Drone on Windows on Arm 64 processors. We actually had no way to build and test the software on these platforms because there was just no publicly-hosted solution that we could use,” he explained.

When they decided to build this solution for themselves, they figured it was going to be useful to other open source projects that also want to ship and support multiple operating systems and architectures. They got Packet on board to offer the infrastructure.

Packet offers a variety of server options with different operating systems and chips, and Rydzewski says this was an important consideration for the open source developers who will take advantage of this service. Packet is making the latest Intel Xeon Scalable, AMD EPYC and Arm-based servers available to users of the Drone Cloud service for free as part of a multi-year donation to support the project.

“As open source software is deployed to increasingly diverse environments, from traditional data centers to cars and even drones, the need for multi-architecture builds has exploded,” Jacob Smith, co-founder and CMO at Packet said in a statement. This is Packet’s way of supporting that effort.

Drone does not intend to establish a paid layer for Drone Cloud, according to Rydzewski, but he hopes it shows what Drone can do, which in turn could attract some developers to the paid version of the product. In addition to supporting the open source version, the company offers a paid version that can be installed on premises as a part of a private cloud-native development environment.

Spinnaker is the next big open source project to watch

Spinnaker is an open source continuous delivery (CD) platform from Netflix and Google, though it now also has the backing of other major software companies. Spinnaker 1.0 launched last July, so it’s not the newest kid on the block, but the service is slowly but surely gaining momentum now, with users that include Target, Adobe, […]

Spinnaker is an open source continuous delivery (CD) platform from Netflix and Google, though it now also has the backing of other major software companies. Spinnaker 1.0 launched last July, so it’s not the newest kid on the block, but the service is slowly but surely gaining momentum now, with users that include Target, Adobe, Daimler and Capital One, as well as a growing ecosystem of vendors who support it.

Today, after a few years of working on the project without any formal structure in place, the Spinnaker project announced that it is growing up and putting a formal governance system in place at the project’s second community summit in Seattle this week.

Like Kubernetes, which has become the de facto standard for container orchestration tools, Spinnaker could become the same kind of standard for continuous delivery. That space, though, already has plenty of incumbents and established players, so chances are this will be a bit more of an uphill battle. Spinnaker and Kubernetes make for a pretty obvious pairing, though, so there’s now also plenty of Kubernetes startups that are looking at how they can best combine the two.

What’s most important in the short run, though, is that Spinnaker is now getting a formal governance structure whereas before, it was basically run as a GitHub project with a benevolent dictator in place.

“That’s something that the community’s been looking for in terms of how do people get a seat at the table,” Netflix’s Director of Delivery Engineering Andy Glover, who oversaw the development of Spinnaker, told me. “The project has largely been run by Netflix and Google. We’ve taken any questions from the community and big companies, whether they be Cisco or Target, about trying to figure out ‘what’s the deal here?’ How do we how do we reduce risk, how do we guard ourselves from Netflix closed-sourcing it or Google’s deciding to license it or something like that.”

So going forward, the project will have a technical steering committee and a steering committee. For now, those committees are staffed with Netflix and Google engineers, but the plan is to open it up to third parties as well. The new governance policy also outlines how developers can start committing code to the project.

In the early days, having Glover and others shepherd the project informally was just fine. Now that the community is growing, though, and more large companies are starting to use Spinnaker, Glover admits that to scale the project, others have to step in. “At Netflix, we tend to do a lot of experimentation without worrying too much up front,” he told me. “Let’s just run fast and see what happens. And with respect to Spinnaker, that was very much run the same way. We said we’d cross that bridge when we get to it and obviously, we got to that bridge a little while ago.”

One thing a lot of people have been wondering about is whether Spinnaker will eventually land at any of the major open source foundations like the Linux Foundation, the OpenStack Foundation or the Apache Foundation. Glover noted that this move is meant to set the stage for that.

Boris Renski, the founder of Mirants, which has recently made a major bet on Spinnaker, tells me that this new governance policy is very much needed (and he’d prefer the project to land with the OpenStack Foundation). He told me that today’s Spinnaker, without formal governance in place, isn’t always the most community-friendly place to be.

“Spinnaker has all the chances to become the de facto continuous delivery tool,” he told me. Putting the governance in place is only a first step, though, Renski actually believes that one of the challenges for the project is the fact that Kubernetes is already putting many of the CD tools for its community in place. Kubernetes, he argues, is suffering from “an OpenStack syndrome” where it has “its fingers in everything” (though to be fair, OpenStack has paired its efforts back quite a bit in recent years). That, he thinks, is not a healthy dynamic and he believes that more specialized tools are the way to go. But Kubernetes is the hot new thing right now and developers are gravitating to it. Yet CD solutions that only cater to Kubernetes discount that most enterprises will want to be able to deploy to other targets, too. Spinnaker, he argues, should be a friend to Kubernetes developers but still remain flexible enough to work for everybody.

He also noted that one problem with today’s Spinnaker community is that it’s mostly driven by users who are trying to solve a near-term tactical problem. “Those users don’t have time and bandwidth to solve longer-term, community-type problems,” he said. What the project still needs in his view is real “pluggability,” that is, the ability to extend Spinnaker and more easily integrate it with third-party systems.

Google, Microsoft and Amazon now back the project and it runs on their clouds. Pivotal, too, recently announced increased support for it and so are many other players in the continuous integration and delivery ecosystem. Pete Erickson, who organized this week’s Spinnaker Summit, told me that he’s expecting about 400 participants from 16 countries and 275 companies at the event. And Glover also noted that about 30 percent of attendees are new to Spinnaker and are simply attending to learn about it and how to bring it to their companies.

JFrog lands $165 M investment as valuation jumps over $1 billion

JFrog wants to change the way we deal with software updates. Instead of large numbered updates you have to manually download, it sees a future of continuous delivery where software is delivered as binaries and updated in the background. Investors must like that vision very much because they showered the company with a $165 million […]

JFrog wants to change the way we deal with software updates. Instead of large numbered updates you have to manually download, it sees a future of continuous delivery where software is delivered as binaries and updated in the background. Investors must like that vision very much because they showered the company with a $165 million Series D investment today, which the company reports pushes its valuation past the billion dollar mark.

The round was led by Insight Venture Partners, and as part of the deal Insight’s co-founder and managing director, Jeff Horing will be joining the JFrog board. Other investors joining the round included new investors and Silicon Valley Funds, Spark Capital and Geodesic Capital, as well as existing investors Battery Ventures, Sapphire Ventures, Scale Venture Partners, Dell Technologies Capital and Vintage Investment Partners. Today’s investment pushes the total invested to-date to over $226 million.

What the company has done to justify this kind of investment is offer a series of products that enable customers to deliver code in the form of binaries. That in turn allows them to deliver updates on a regular basis in the background without disturbing the user experience. In a world of continuous delivery, this approach is essential. You couldn’t deliver multiple updates a day if you had to take down your service every time you did it.

The JFrog platform is actually made up of multiple products, but the main one is JFrog Artifactory where companies can add the latest binaries (updates) and deliver them to customers in the background. It’s not unlike, GitHub, but whereas GitHub is a repository for downloading software and updates, the Artifactory is a place to deliver these updates automatically without user involvement. It also handles other DevOps functions like security, access control and distribution.

JFrog platform. Diagram: JFrog

CEO and co-founder Shlomi Ben Haim was happy to reveal that the company’s valuation had entered unicorn territory, but he wasn’t willing to share an exact number. “I don’t want to get into details, but we exceeded the billion dollar valuation. We are north of $1 billion already and we are building the company to generate the revenue to justify it,” he told TechCrunch.

He wasn’t discussing specific revenue numbers, but reports the company has a goal of a billion dollars in revenue by 2025, and he says they are working toward that. He did say they have had 500 percent revenue growth since the $50 million round in 2016, and that they tripled the number of employees to 400, while doubling the number of products they offer. They currently have 4500 customers including 70 percent of the Fortune 100.

So fair to say things are going well for the company. Ben Haim says the ultimate goal for the company is to deliver software in the background for scenarios like your operating system or your Tesla. Instead of shutting down your car or computer for the next software update, it will just happen over the air in the background. We are obviously a ways from fulfilling that vision, but investors are clearly betting on that potential.

JFrog lands $165 M investment as valuation jumps over $1 billion

JFrog wants to change the way we deal with software updates. Instead of large numbered updates you have to manually download, it sees a future of continuous delivery where software is delivered as binaries and updated in the background. Investors must like that vision very much because they showered the company with a $165 million […]

JFrog wants to change the way we deal with software updates. Instead of large numbered updates you have to manually download, it sees a future of continuous delivery where software is delivered as binaries and updated in the background. Investors must like that vision very much because they showered the company with a $165 million Series D investment today, which the company reports pushes its valuation past the billion dollar mark.

The round was led by Insight Venture Partners, and as part of the deal Insight’s co-founder and managing director, Jeff Horing will be joining the JFrog board. Other investors joining the round included new investors and Silicon Valley Funds, Spark Capital and Geodesic Capital, as well as existing investors Battery Ventures, Sapphire Ventures, Scale Venture Partners, Dell Technologies Capital and Vintage Investment Partners. Today’s investment pushes the total invested to-date to over $226 million.

What the company has done to justify this kind of investment is offer a series of products that enable customers to deliver code in the form of binaries. That in turn allows them to deliver updates on a regular basis in the background without disturbing the user experience. In a world of continuous delivery, this approach is essential. You couldn’t deliver multiple updates a day if you had to take down your service every time you did it.

The JFrog platform is actually made up of multiple products, but the main one is JFrog Artifactory where companies can add the latest binaries (updates) and deliver them to customers in the background. It’s not unlike, GitHub, but whereas GitHub is a repository for downloading software and updates, the Artifactory is a place to deliver these updates automatically without user involvement. It also handles other DevOps functions like security, access control and distribution.

JFrog product flow

JFrog platform. Diagram: JFrog

CEO and co-founder Shlomi Ben Haim was happy to reveal that the company’s valuation had entered unicorn territory, but he wasn’t willing to share an exact number. “I don’t want to get into details, but we exceeded the billion dollar valuation. We are north of $1 billion already and we are building the company to generate the revenue to justify it,” he told TechCrunch.

He wasn’t discussing specific revenue numbers, but reports the company has a goal of a billion dollars in revenue by 2025, and he says they are working toward that. He did say they have had 500 percent revenue growth since the $50 million round in 2016, and that they tripled the number of employees to 400, while doubling the number of products they offer. They currently have 4500 customers including 70 percent of the Fortune 100.

So fair to say things are going well for the company. Ben Haim says the ultimate goal for the company is to deliver software in the background for scenarios like your operating system or your Tesla. Instead of shutting down your car or computer for the next software update, it will just happen over the air in the background. We are obviously a ways from fulfilling that vision, but investors are clearly betting on that potential.