Cloud Native Computing Foundation KubeCon + CloudNativeCon North America 2020 - Virtual, 4 Dec 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Keynote: CNCF Projects Update - Constance Caramanolis, KubeCon + CloudNativeCon NA 2020 Co-Chair

Description

Don’t miss out! Join us at our upcoming event: KubeCon + CloudNativeCon Europe 2021 Virtual from May 4–7, 2021. Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.

Keynote: CNCF Projects Update - Constance Caramanolis, KubeCon + CloudNativeCon North America 2020 Co-Chair & Principal Software Engineer, Splunk

https://sched.co/eoEF

A

Hi everyone, I'm constance, karanilis and I'll, be doing your cncf project updates. Today before we get started, um I want to talk a little bit about something we're changing this time, trying to see how it goes as you'll notice during this kubecon cloud native con there's an increased focus on telling end user stories and we're going to continue that theme here today during project updates.

A

Let's get started our first project we're going to talk about is falco.

A

Falco is a runtime security project and it consumes signals from the linux kernel and container management tools such as docker and kubernetes falco parses these signals and asserts them against security rules. If a rule is violated, an alert is triggered. Now. Let's look at some of the cool statistics that have happened in the past year, one there have been over 700 000 downloads of the kernel module in 10 days.

A

That's an amazing number 60 new contributors this year and these contributors come from companies like ibm, coveo, shopify, aws, app dynamics, inno team really wide variety of contributors. That's amazing. There are 12 adopters and many independent users.

A

Now, what have been some updates to the project? One is a user space driver additions, such as support for fargate workloads using pdig support for rm64 arm7.

A

Yes, that means lower energy processing stability, slash improvements to the falco engine for better performance.

A

Additional community integrations such as falco sidekick and a growing number of additional community integrations and there's been some user adoption improvements, such as improved helm, charts, falco, ctl focal cuddle. I'm going to need someone to correct that for me afterwards, falco explorer for prometheus and improving bent generators event generators.

A

I really like the callout here of better user adoption improvements right. It's always about making sure that it's easy for people to get on board, and this is really wonderful to see in terms of end user. I will not be able to do justice to shane lawrence and chris nova.

A

If you didn't have a chance, I highly suggest for you to look to watch their keynote from eu shane lawrence comes from shopify and he describes what uh how shopify uses falco to give you the tldw. The too long didn't watch, shopify uses falco and its rule sets in production. For detection and prevention it does things such as detecting data, egress example, coin mining, metadata server abuse and a remote code. Execution output.

A

Please check out their keynote. They will do much better description of how things are done. Thank you, uh shopify shane lawrence and chris novo. Also for doing kubecon. U, and thank you falco on to our next project. Thanos thanos is a prometheus-based scalable time series database that provides a global query view. High availability and data backup with historical, cheap access as its core features. All in a single binary.

A

They, it is three years old, they have 281 contributors and direct quote from the maintainers. They are very grateful for that. Thank you. Contributors.

A

There are 6 500 stargazers and in 2020 they focused on mentoring. They had nine students. These students came from programs like cncf community bridge and google summer of code, and they had a really positive impact on the project. I'll mention one of their additions in a minute and they have high commit velocity.

A

Over the past year, they've been averaging six commits per day. This is amazing. uh This is a chart here on the right that I want to see for everyone. um This is just great work, everyone.

A

What about some updates? So uh we need to give thanos a round of applause, that virtual round of applause and slack everyone.

A

It has reached incubation status in april 2020., congratulations, thanos community, that is great work, they've added support for native response, caching, collaboration with cortex project. Another thing I want to highlight collaboration with other projects. This makes me really happy to see it is you know we are all providing solutions for the cloud native world and seeing this collaboration just makes me really happy.

A

They've added new react uis to all components. This is also a really big thank you to the community and to the mentees.

A

There is a heavy focus on analytics. Olap integrations through obslytics community subproject.

A

There's offer full multi-tenancy. This allows for flexible deployment of various read storage and right tenancy and the rule of limitation from courier prom, ql, single cpu, concurrency and prometheus. It means it's multi-core. It has sped up querying.

A

Thank you, thanos.

A

Let's talk about rook. Now we have to give rook a round of applause. They graduated earlier this year.

A

Rook provides an open source storage solution for kubernetes through operators and crds it automates management of storage providers, including nfs cassandra, cockroachdb and hugobyte db.

A

There are over 7 800 stargazers 295 contributors uh that one want me to say again. Thank you to the rick community maintainers over 180 million downloads. That's amazing and most recently, um there's a release of the 1.5, and it contains improvements such as support for stretch, clusters, encryption with kms and mirroring block data across clusters.

A

Now the next end user, I'm really excited about mostly because it's not you know, we don't really hear so much about research and education use cases.

A

So uh the pacific research platform project is a partnership across many many institutions, both at the government level, national science foundation, department of energy and also multiple research, universities in the united states and around the world. So prp aims to create a seamless research platform that encourages collaboration on a broad range of data, intensive fields and projects.

A

Now, naturally, using kubernetes was a good idea for that, so they run a kubernetes cluster, it's mostly located in california, but also other parts of the country, and they have some nodes in the pacific asian region and europe. All of this is connected uh through a with a fast 100g research network.

A

Now this cluster incorporates four stuff clusters managed by rook two petabytes in the u.s western region, 500 terabytes in u.s eastern region, eight terabytes, video wall, ssd only local 130, terabytes, nvme, only local and soon to be added in asia, pacific pool now um the following is a direct quote from prp about how rook has enabled them to be successful.

A

Rook is tremendously helping us to deploy a reliably, a reliable, fast store to scientific users. Rook makes administration of large ceph installations easier and removes multi multiple tasks of nodes management from personnel which helps reduce costs and increase efficiency. Data storage is a key component of scientific compute, computations and rook. Does it perfectly for us?

A

Thank you, prp and thank you. Rook for sharing.

A

On to the last project, we're going to talk about today is the tests. The test is a database solution for deploying scaling and managing large clusters of open source database instances. It supports mysql in mariadb.

A

It is architected to run as effectively on public or private cloud architecture as it does on dedicated hardware, and it combines and extends many important sql features of the scalability of nosql database, and so some of the things problems you know that it tries to solve are scaling sql database by sharding it migrating from bare metal to private, private or public cloud, not for live, but, as one word was trying to say and deploying and managing a large number of scale sequel database instances.

A

One fun fact that I know about before is that the tesla's actually running on kubernetes at youtube before kubernetes 1.0, it's been around for a while. It's awesome in terms of some updates. There have been four major releases this year and the latest release. 8.0 is compatible with over 10 modern sql frameworks, including wordpress and ruby, on rails, active record and for a last end, user story slack.

A

So, at the beginning of 2017 slack began adoption of the tests by in august of 2017, it started serving production traffic after completing a re-architecture and migration of slack's core product features by early 2020, roughly 70 of traffic was passing through the tests and they had designed a framework to migrate one table at a time now, you're, probably wondering what about the remaining 30, and that is the crux of the problem.

A

They had done a calculation that it would take three to five years, or at least four engineers to migrate the remaining thirty percent of the legacy system, and that's a really long time now um they knew it was going to take a while and so in 2019. They spent effort uh investigating alternative alternative strategies to do the migration and what they found was if they're leveraging the view replication functionality is that they can migrate entire clusters to the tests.

A

Now this resulted in as of to now the 99 of traffic at slack, uses the tests and that the migration will be complete by end of year.

A

And just a shared direct quote: we moved hundreds of terabytes of data over a thousand, my sequel, shards zero downtime, not a single outage, equally important, it was transparent to our applications.

A

Thank you slack for sharing and thank you to tess.

A

Now before we cut off for a moment, I want to call it some other graduations, tick: v, tough, the update framework, harbor and helm. These are other projects that have graduated since we all met together in san diego last year. Please give them a virtual round of applause, use those slack emojis. Please congratulations to the community contributors maintainers a lot of hard work and it's well earned and our last last message of the day.

A

Please go to cncf dot, io projects to get a full list of all the different projects that are supported and also use leverage to maintain your track session to meet the maintainers. I'm sure you probably have a lot of questions, and so please use this time to reach out to find out what's going on with the project and ask you know, get answers to your questions. Thank you very much everyone and have good coupon. Bye.