Kenneth Goodwin

As you have probably seen in earlier posts, the preview version of MariaDB Vector is out and ready for you to play with. We have had input from several different places during the development of this feature. This, of course, includes hardware manufacturers such as Intel Corporation. In the background, #Intel have been prototyping using #AVX512 instructions for dot product and bloom filter. Both of these are functions are part of vector searches. https://lnkd.in/d3K8xJet #mariadb #database #vectorsearch #mariadbvector

17

What can evolutionary architecture look like in #serverless on #AWS? 🧠 In this video I talk through ✔️ my own personal approach with clean code ✔️ show examples of what bad looks like ✔️ and an example of swapping services or going from functions to containers easily with minimum of fuss. Check out this article for full code example in GitHub: https://lnkd.in/eHdXJ7hu I hope you enjoy it 💜

38

7 Comments

Linux Professional Institute (LPI) is celebrating 25 years promoting the use of #opensource by supporting the people who work with it. 🎂🐧🎈🌍 We couldn’t have done it without our community: as thanks, every 25th candidate who takes an #LPI exam Oct. 1 - Dec. 31st, gets a free voucher for their next exam! Learn more: https://lpi.org/gvbb #LPI25 #LPIC #LinuxEssentials #Linux #opensource #FOSS #WebDevelopment #cybersecurity #DevOps #BSD #programming #coding #SQL #Javascript #WDE #OSE #sysadmin #Linuxcertification

503

7 Comments

Managing variables in Terraform is essential for flexibility and security in infrastructure automation. Terraform offers several methods to define and prioritize variables, crucial in dynamic environments like CI/CD pipelines.

7

If you want your Linux to be secure, you should use a distro based on a long-term support kernel.

4

Did anyone else know that 32 bit UNIX time won't work as of 1/19/2038? It seems this is bigger than Y2K scare! From: https://lnkd.in/geKj-J2C What happens on January 19, 2038? On this date the Unix Time Stamp will cease to work due to a 32-bit overflow. Before this moment millions of applications will need to either adopt a new convention for time stamps or be migrated to 64-bit systems which will buy the time stamp a "bit" more time.

6

10 Comments

An embedding subsystem is one of four subsystems needed to implement #RAG. In this post, Keith Pijanowski builds a distributed embedding subsystem that can run on engineering workstations and in a fully distributed #cloudnative production environment. MinIO is the best storage solution for generative #AI—embedding models need a high-speed, scalable storage solution. https://hubs.li/Q02L29ly0

11

1 Comment

An embedding subsystem is one of four subsystems needed to implement #RAG. In this post, Keith Pijanowski builds a distributed embedding subsystem that can run on engineering workstations and in a fully distributed #cloudnative production environment. MinIO is the best storage solution for generative #AI—embedding models need a high-speed, scalable storage solution. https://hubs.li/Q02L2bdL0

9

An embedding subsystem is one of four subsystems needed to implement #RAG. In this post, Keith Pijanowski builds a distributed embedding subsystem that can run on engineering workstations and in a fully distributed #cloudnative production environment. MinIO is the best storage solution for generative #AI—embedding models need a high-speed, scalable storage solution. https://hubs.li/Q02L29nn0

13

An embedding subsystem is one of four subsystems needed to implement #RAG. In this post, Keith Pijanowski builds a distributed embedding subsystem that can run on engineering workstations and in a fully distributed #cloudnative production environment. MinIO is the best storage solution for generative #AI—embedding models need a high-speed, scalable storage solution. https://hubs.li/Q02L2bcP0

11

NetBSD Updates Guidelines to Prohibit AI-Generated Code - https://lnkd.in/gRUe83J8 NetBSD, the popular UNIX operating system, has updated its commit guidelines to ban the use of code generated by AI.

1

As a veteran of many language wars, as well as someone who participated in a project to create a Linux-compatible OS layer in Rust, I have a lot of sympathy for the protagonists of this story. The fact is, no matter what you do, a huge portion of what it takes to bring a new language platform to a project is going to be nontechnical nonsense. The existing maintainers, who already have a hard job, get very concerned when they see anything that might make that job harder. You have to start with modest goals, and communicate them clearly and carefully, making sure that those folks don't feel like you are asking them to do anything that they don't see as necessary. It will be a very long time before a core team are willing to support anything you do, and in the meantime, you have to do all of the heavy lifting - no one is going to help you. Java was about 7 years into its run at Google - and services like Gmail and Ads were already written in it - before teams that owned basic services and libraries like RPC and protobuf accepted that Java customers were a priority and were willing to devote resources to supporting them. Before that, those libraries were supported by the customers, or by volunteers, or - sometimes - by no one. The best approach I've seen seems to be to find one or two customers who are excited about your platform, and to work with them on everything they need to make them a success. That was Go's initial path within Google. And even with modest, specific initial goals, you are going to face everything from open embrace to mild skepticism to outright hostility. You have to face it all with a smile. You have to answer even the most hostile questions in a constructive and positive way. You have to patiently work through lists of dozens of requirements, many of which may seem pointless. You have to build up success stories, and relationships with the developers and leads, and make sure everyone understands that you are an integral part of the community. If you build consensus, show up in good faith, and deliver on what you've promised, the hostility subsides. And those relationships you've built pay off when those teams are finally ready to invest in your work. It's not for everyone. It's especially hard when you have people who want a new language platform to succeed because they can see benefits for everyone on one side, and 30MLOC of C maintained by domain experts who have been successful at what they are doing for decades on the other. If I were doing it, I'd probably focus on getting a single production-ready driver out the door. Drivers are notorious sources of memory errors in the Linux kernel (for a variety of reasons), and third parties may be the most interested in Rust adoption. I see they have a non-production-ready driver in their fork. It will be interesting to see where they go next. #rust #linux https://lnkd.in/gWjW3RPF

34

30 Comments

Have you heard about Kyverno? It's a powerful policy engine designed specifically for Kubernetes. With Kyverno, you can: - Validate, Mutate, and Generate - Kubernetes resources effortlessly. - Secure your software supply chain by verifying OCI image signatures and artifacts. - Manage policies using familiar tools like `kubectl`, `git`, and `kustomize` - no need to learn a new language! Kyverno integrates seamlessly into your CI/CD pipeline, ensuring compliance, security, and automation at every step. 🌟 Check it out on GitHub: [Kyverno](https://lnkd.in/dguXtbFP)   Learn more: [Kyverno.io](https://kyverno.io/) #Kubernetes #DevOps #CloudNative #PolicyManagement #Kyverno 🔗 Let's connect and discuss how Kyverno can enhance your Kubernetes workflows!

1