• Kubernetes in Production

    Microsoft Reactor

    **IMPORTANT: You must register here to have a confirmed ticket for the event: https://www.skipcard.com/events/5d24daacc8107e0004c2854d?utm_medium=meetup&utm_source=DevOps-Exchange-SanFrancisco --- Here is an opportunity for you to exchange some of the latest ideas and technologies for running online services through 2-3 short presentations by invited speakers, and a chance to chat it after over drinks/snacks with new friends. The event features three talks: 1. ManagedKube CEO & Co-founder on "Kubernetes in Production" 2. Spotinst Founder & CEO on "Serverless Containers" 3. Kong on "Mutual TLS with Kong: Secure your services at scale" ---- Details: Garland Kan - ManagedKube CEO & Co-founder Title: Kubernetes in production Interested in how Kubernetes can help you deploy, scale, and manage your application? Wondering why Kubernetes is so great and has grown so quickly since its launch in 2015? Need help running or converting your app on Kubernetes? Come learn that and more at evening talk with Q&A, Get Started with Kubernetes, with Garland Kan! About the speaker: Garland Kan is a DevOps software engineer that helps small to medium-sized start-ups run large-scale, reliable applications. He works with the entire development team architecting, designing, building, optimizing, and operating infrastructure in the cloud (AWS, Google Cloud, and Azure). --- Amiram Shachar - Spotinst Founder & CEO Title: "Serverless Containers" - Nodeless Kubernetes & Vertical Pod Autoscaling Serverless Containers or "Nodeless" Kubernetes, is the future of Containers infrastructure. Matching and scaling the right infrastructure to ever-changing micro-services deployments is a challenge. In this talk, we will review the evolution of Containers Auto Scaling in Kubernetes both Horizontal and Vertical, discuss the trade-offs, and introduce a new approach to deploy Serverless Containers in a "Nodeless" cluster. No Infrastructure or VM to manage, just Containers that can self-optimize. The session includes a live demo of Ocean. About Spotinst: Spotinst offers a SaaS optimization platform that enables companies of all sizes, major cost reduction and availability across all cloud providers. Spotinst optimizes your infrastructure for performance, availability and the lowest cost. --- Datong Sun - Systems Engineer at Kong's Core Team. OpenResty project core maintainer. Datong will talk about Mutual TLS authentication, use cases, pros/cons and best practice while deploying mutual TLS secured infrastructure. Datong will also showcase the mutual TLS support that has recently been introduced in Kong 1.3.

  • Manage and Optimize Kubernetes within your Hybrid Cloud Infrastructure

    Hello SF Kubernauts! Join us for a technical deep-dive into the world of Kubernetes within Hybrid Cloud together with Joey Yep, Solutions Architect @ Spotinst and Kamesh Pemmaraju, Head of Product Marketing @ Platform 9 Matching and scaling the right infrastructure to ever-changing micro-services deployments is a big challenge to many organizations. When it comes to Kubernetes, there are too many things an operator should take care of and configure such as selecting the best instance type and size, determining when to scale up or down, making sure all of the containers are scheduled and running on the best instances and that is even before starting to think about cost resource optimization. Agenda: 5:30 pm - 6:00 pm: Gathering, pizza and beers 6:00 pm - 6:40 pm: 3 layers approach of automating and optimizing containers workloads // Joey Yep, Spotinst In this session, we will demonstrate and introduce a new approach, provided by Spotinst Ocean to this challenge by using the 3 layers approach of automating and optimizing containers workloads: 1. The pricing model layer (Spot, On-Demand & Reserved Instances). 2. The instance sizing layer - choosing the right infrastructure size & type to satisfy the actual containers requirements. 3. The containers utilization layer - changing in real time the limits and resource requests of the containers based on their consumption and utilization. 6:40 pm - 7:20 pm: Top Micro-services, open source trends and recommendations for 2019 // Kamesh Pemmaraju, Platform 9 The cloud computing landscape is undergoing yet another major inflection point. The public clouds continue to explode adding more and more services and expanding their global footprint. In the meantime, open source momentum has picked up considerably for containers, Kubernetes, and related services and this is expected to be a dominant force in on-premises private cloud environments. Join us to learn about the top four trends for 2019 and recommendations on how organizations can navigate through these trends to prepare themselves for the present and future of cloud-based infrastructure management.

  • Beyond Kubernetes Cluster Autoscaler

    San Francisco

    Hello SF Kubernauts! Get ready for our first meetup hosted in Spotinst new SF office! Come and join a technical deep-dive into the world of Kubernetes Autoscaling featuring Trevor Bossert, Manager of DevOps @ Primer and Amiram Shachar, Founder & CEO @ Spotinst. In this Meetup, we will discuss the design considerations for scaling Kubernetes clusters, such as pod resource requirements (CPU, MEM, GPU, Custom resources) and pod dependencies (Volume Claims, Affinity, Node-selectors, Tolerations). We’ll also present our custom cluster autoscaler that implements these principles. Agenda: 18:00 - 18:30: Gathering, pizza and beers 18:30 - 19:00: Autoscaling Kubernetes Clusters - Amiram Shachar// Spotinst 19:20 - 20:00: More than Autoscaling Trevor Bossert// Primer- *Live Demo* About our Speakers: ----------------------------------- Amiram Shachar, CEO & Founder of Spotinst A hands-on technical leader with significant experience leading DevOps and software engineers to build and maintain large-scale cloud architectures and infrastructure. He started his career in the Israeli military - managing and operating the IDF's data center. As part of his role, he implemented VMware and virtualization for the first time ever in the IDF. Trevor Bossert, Manager of DevOps @ Primer Is an accomplished DevOps Engineer with well over a decade of experience. He is passionate about removing obstacles that impede developers. Why have 800 little scripts and processes that someone has to implement chunk by chunk when you could have a system that flows on its own- you could even say, automated? These are the things that keep Trevor up at night :)

    6
  • Amazon EKS - Everything you need to know

    WeWork California St.

    Amazon EKS is generally available for all AWS customers. Thinking about using EKS in production or integrating EKS with your already existing Kubernetes tools? In this Meetup we will present everything that you need to know in order to provision and use Amazon EKS in production. Join AWS and Spotinst for this Meetup to learn: *Advanced Introduction to Amazon EKS* We will dive deep into how EKS is built and how it leverages multiple AWS services to eliminate a single point of failure and increase security. *The Power of Running EKS Worker Nodes on a blend of EC2 Spot & Reserved Instances* We will present a live demo of an EKS cluster running on EC2 Spot Instances and explain how Kubernetes and containers are a natural fit for EC2 Spot. 18:00-18:20 : Gathering & Networking 18:30- 18:55: Anatomy of Kubernetes and Amazon EKS Speaker: Shawn OConnor, Solution Architect at AWS 19:05-19:30: Demo - Create Amazon EKS and deploy your first service behind an ELB and Persistent Volumes (PVC). Speaker: Shawn OConnor, Solution Architect at AWS 19:40-20:05: Demo: How to blend On Demand & Reserved & Spot Instances in a single EKS cluster. Speaker: Amiram Shachar, Founder & CEO at Spotinst

    6
  • Running Containers at Scale

    Pubnub

    Please join Spotinst & Cloud Mafia for a night of networking, tech talks and some great food and drinks! Agenda: 6-6:45pm Networking & food/drinks 6:45-8pm Tech talks 8-8:30pm Networking *PubNub will be hosting and sponsoring this event. Topic: Autoscaling Kubernetes and ECS with Maximum Efficiency TALK#1 Speaker: Orr Weinstein is the VP of Product at Spotinst, a company that helps enterprises save up to 80% on their cloud computing costs. Orr has 10+ years of experience in software development and product management, and is the former product manager for AWS Lambda. Abstract: Scaling a Kubernetes cluster (effectively) can be quite a challenge, especially when it comes to heterogeneous environments (made up from different machine types, and sizes). To increase the cluster’s efficiency, choosing the right instance size and type for the right workload is crucial. In this Meetup, we will talk about two important concepts of k8s autoscaling; “Headroom” & “2 Levels Scaling” that are crucial to understand in order to scale k8s efficiently. TALK#2 Speaker: Dan Garfield is a full-stack engineer at Codefresh, a DevOps platform designed for containers and Kubernetes. Codefresh includes CI/CD, image management, and a Kubernetes dashboard for managing releases. Dan is a Kubernaut and old school hacker who likes to build robotic-chicken coops in his spare time. Abstract: Bringing your full-application stack together to fully validate changes can be a real challenge if you don't have a scalable scheduler like Kubernetes. In this talk, we'll show the best-practices for using Kubernetes to spin up environments on-demand as well as some of the validation possible when you have you have everything available. It will include an intro to GitOps, package management, and automation. Please RSVP: This event is free to attend. Food and drinks will be provided. You must RSVP.

    1
  • Big Data on Serverless

    WeWork California St.

    6:00pm Pizza, drinks, and networking 6:30pm First talk: Running Big Data workloads on Serverless In this talk, we'll talk about Big Data patterns as a service. We will present a MapReduce reference architecture demonstrates how to use Serverless functions in conjunction with Amazon S3 to process data stored in S3. By leveraging this pattern, you can build a cost-effective pipeline to run ad hoc MapReduce jobs. The price-per-query model and ease of use make it very suitable for data scientists and developers alike. Speaker; Kevin McGrath: Currently VP of Serverless Architecture at Spotinst, Kevin was previously the Senior CTO Architect at Sungard AS where he led the vision and future of architectural strategy for 5 years after spending the previous 10+ years as a Senior Architect at AT&T. ----5 minute break---- 7:05pm Demo: Serverless MapReduce Hands on example of a real Mapreduce operation completely running Serverlessly. Jeffrey Noehren, Serverless Developer at Spotinst and Kevin McGrath VP of Architecture, Spotinst.

    3
  • Creating a ridiculously cost-effective data pipeline! (SV edition)

    We'll be joining up with our friends at Microsoft Data Platform SV to teach how you can run the most cost-effective data pipeline on the planet by running it on Spot. * Network with other Developers and DBAs * Get the latest and greatest SQL Server news * Have your SQL Server questions answered by Microsoft insiders * Get the inside track on emerging technologies * Learn valuable information for career development * Enjoy dinner, drinks, and valuable software giveaways 6:30 to 7:00 pm – Introductions, food, networking 7:00 to 8:30 pm – Presentation and Q&A Presentation Topic: Creating a ridiculously cost-effective data pipeline Reducing costs for your data pipeline can be complex but it doesn't have to be. In this talk, we'll explain how you can use spare capacity (eg Spot Instances or Low-priority VMs) to reduce your data pipeline costs by up to 80%. All cloud providers have spare capacity that they sell for heavily discounted prices, but this excess capacity comes with a major caveat- you'll be evicted from these instances at two minutes notice once the spare capacity is requested by a customer paying full price. We'll dive into what this spare capacity is, how to use it, and how you can use this spare capacity to save 80% not only on transient applications like dev environments but also production workloads with 100% guaranteed availability.

  • Creating a ridiculously cost-effective data pipeline!

    Microsoft Reactor

    We'll be joining up with our friends at Microsoft Data Platform to teach how you can run the most cost-effective data pipeline on the planet by running it on Spot. * Network with other Developers and DBAs * Get the latest and greatest SQL Server news * Have your SQL Server questions answered by Microsoft insiders * Get the inside track on emerging technologies * Learn valuable information for career development * Enjoy dinner, drinks, and valuable software giveaways 6:30 to 7:00 pm – Introductions, food, networking 7:00 to 8:30 pm – Presentation and Q&A Presentation Topic: Creating a ridiculously cost-effective data pipeline Reducing costs for your data pipeline can be complex but it doesn't have to be. In this talk, we'll explain how you can use spare capacity (eg Spot Instances or Low-priority VMs) to reduce your data pipeline costs by up to 80%. All cloud providers have spare capacity that they sell for heavily discounted prices, but this excess capacity comes with a major caveat- you'll be evicted from these instances at two minutes notice once the spare capacity is requested by a customer paying full price. We'll dive into what this spare capacity is, how to use it, and how you can use this spare capacity to save 80% not only on transient applications like dev environments but also production workloads with 100% guaranteed availability.

  • Azure Low-priority VMs: the easiest way to save 85% on Azure compute costs

    Microsoft Azure has recently (and finally!) released its new excess capacity weapon - Low-priority VMs (aka Spot Instances) that can reduce compute costs by 80-90%! With Azure LPVMs You can dramatically improve your application performance and reduce your infrastructure costs while choosing the right workload. In this meetup, Microsoft Solutions Architects and Spotinst will share ideas and insights that demonstrate how to enable the use of Low-priority VMs for testing and production workloads, the major benefits, the significant cost reductions and how to get started - best practice. With over 2 years of experience in running workloads on Cloud providers excess capacity, Spotinst runs everything from testing Dev environments to mission-critical applications for its clients in more than 30 countries around the world. Agenda: * 6:15 PM - 6:45 PM: Pizza, drinks, and networking * 6:45 PM - 7:10 PM: Introducing Azure Low-priority VMs - save 85% on compute costs Paulo Renato - Principal Solutions Architect, Microsoft. * 7:10 PM - 7:35 PM: Using Spotinst to automatically leverage and orchestrate the use of Low-priority VMs * 7:35 PM - 8:30 PM: Questions, demos, and more chatting