Thursday, February 22, 2018

The Cloudcast #335 - Managing Waste in the Cloud

Brian talks with Andy Richman (Product Manager @ParkMyCloud) about which groups are responsible for cloud costs, how expectations and behaviors are changing as applications move to the cloud (Finance vs. DevOps), do we need DevFinOps, the technology that's making it easier to reduce costs, and tips for driving better cost management.

Show Links:

Show Notes
  • Topic 1 - Introduction of Andy
  • Topic 2 - This survey was focused on “waste” in a public cloud environment. Public cloud was supposed to be less expensive and more agile. Are you seeing people having to reset expectations?
  • Topic 3 - There seem to be two phases of managing costs in the cloud. One is focused buying the right instances at the right time. The other is focused on more day-to-day operations. Are you seeing this sort of separation of functions, and is one a finance function and other a DevOps function?
  • Topic 4 - Let’s get back to “waste”. We went down this road with virtualization many years ago - just turn off what you don’t use. It was the “turning it back on” that actually scared people. What are you learning about more intelligent ways to manage the “turning off/on” process?
  • Topic 5 - AWS has started making more of their services “Ops-less” (sort of), like Lambda, and now Aurora, where you don’t have to think about the infrastructure planning as much. Do you find that people ask why more of the cloud isn’t like this, or they understand that not all workloads run (as before) in these environments?
    Feedback?
    Posted by
    Labels: , , , , , , , , , ,

    Thursday, February 15, 2018

    The Cloudcast #334 - The Future of Edge Computing

    Brian talks with Derek Collison (@derekcollison, Founder and CEO at Synadia Communications) about the future of edge computing, the impact of AI/ML on edge systems, and how Telcos and open source communities will evolve with edge computing opportunities.

    Show Links:

    Show Notes
    • Topic 1 - Welcome to the show. We’ve spoken with you many times over the years, but first time on the podcast. Your background is very well documented, so maybe give us a little glimpse into your latest company - Synadia Communications.
    • Topic 2 - (Public) Cloud Computing has grown in lock-step with the growth of the smartphone (since 2007), creating a specific pattern of application/communication and data collection. What are the forces that are driving all the attention around the Edge Computing evolution (e.g. sensors, smart cars, peer-to-peer patterns, etc.)?
    • Topic 3 - What do the economics of Edge Computing look like today? Where are the big bottlenecks, or areas for big disruption? Does it vary widely depending on the application (e.g. sensor data vs. streaming vs. telemetry) , or geography (e.g. dense cities, transportation networks, rural areas, etc.)
    • Topic 4 - You’ve talked about the Google Brain project for a long time. We’ve seen growth around centralized AI/ML for the last 5-7 years. Do you expect a different mode of AI/ML that has to emerge because Edge Computing patterns will be significantly different than centralized models, or do they need small adaptations?
    • Topic 5 - You’ve need heavily involved with messaging technologies over the years, from TIBCO to NATS. Can you share some insight into how messaging (or async communications) that might need to evolve at the edge, in more distributed types of systems?
      Feedback?
      Posted by
      Labels: , , , , , , ,

      Thursday, February 8, 2018

      The Cloudcast #333 - Infrastructure 3.0 for AI and ML

      Aaron and Brian talk with Lenny Pruss (@lennypruss, Partner at Amplify Partners) about the evolution of application and infrastructure architectures, how AI/ML are radically changing how applications are designed, the new inputs to application systems, and how VCs are investing in companies that can augment new cloud services.

      Show Links:

      Show Notes
      • Topic 1 - Welcome to the show. Tell us about your background and how you eventually got into the VC side of the business.
      • Topic 1a - Infrastructure and applications evolve in 'lock-step', so where are we today in terms of this evolution and where are we going?"
      • Topic 2 - You mention this concept of “connectivity vs. cognition” is what makes ML/AI radically different from software of previous generations. Let’s discuss what that means.
      • Topic 3 - “The result is an application that is “smart,” but exceptionally data-intensive and computationally expensive”. What are the problems this creates on today’s infrastructure?
      • Topic 4 - “For ML/AI to reach its full potential, it must graduate from the academic discipline it is today into an engineering one. What that means in practice is that there needs to be new abstractions, interfaces, systems, and tooling to make developing and deploying intelligent applications easy for developers.” - Give us some of the first steps that will make this happen.
      • Topic 5 - As a VC, you’re in the business of finding the companies that might turn this into a reality. Who are some of those companies that are on your radar?
        Feedback?
        Posted by

        Friday, February 2, 2018

        The Cloudcast #332 - A Cloud Perspective from Israel

        Brian talks with Ariel Munafo (@ArielMunafo, TotusCloud) and Amit Dunsky (@amit_dunsky, high-T Cloud Computing), co-hosts of the CloudTalk Podcast about the Cloud Computing market in Israel. This is a co-branded show and will also be published on CloudTalk podcast channel.

        Show Links:

        Show Notes
        • Topic 1 - Welcome to the show. Tell us about your background in Cloud Computing and what drove you to create your podcast?
        • Topic 2 - Let's talk about the differences in cloud cultures between the US vs Israel, in terms of venture-capital, adoption, support, innovation, etc.
        • Topic 3 - Let's discuss the regulatory issues affecting cloud adoption in each country.
        • Topic 4 - You both do some work with Israeli government agencies, what about cloud adoption by the government offices and agencies?
        • Topic 5 - Let’s also discuss local tech conferences as an index to how much the cloud is presence and how it indicates the cloud vendors' plans ahead.
          Feedback?
          Posted by
          Labels: , , , , , , ,

          Thursday, January 25, 2018

          The Cloudcast #331 - Has SRE replaced DevOps?

          Aaron and Brian talk with Rob Hirschfeld (@zehicle, CEO @rackngo; Kubernetes Cluster Ops Co-Chair) about the consistency, continuum and confusion between the concepts of DevOps and Site Reliability Engineering (SRE).

          Show Links:

          Show Notes
          • Topic 1 - What is the State of DevOps today? On one hand, there’s Gene Kim’s DevOps Reports (all is great), on another hand is DevOps Days which has become about Empathy, and somewhere in between are companies struggling with all of this silo-busting and automation and constant change. So where are we?
          • Topic 2 - The DevOps community seemed to want to reject all sort of labels and titles (DevOps engineer, DevOps certified, etc.) and how there is this “SRE” (Site Reliability Engineering) concept. Is this just a new name for DevOps?
          • Topic 3 - Like NetFlix had microservices, so everybody needed microservices - Google has SRE, so now everyone needs SRE? How does SRE fit into a non-Google company?
          • Topic 4 - Many Infra/Ops-centric people have been trying to learn automation and some basic programming (e.g. Python, Powershell/Scripting). SREs are often described as programmers that live in the Ops world. Can these current Infra/Ops people evolve to SRE?
          • Topic 5 - Do you find that DevOps or SRE apply more (or less) to using certain types of technologies vs. other technologies?
            Feedback?
            Posted by
            Labels: , , , , , , , , , , ,

            Thursday, January 18, 2018

            The Cloudcast #330 - Oracle’s Next-Generation Cloud IaaS

            Brian talks with Clay Magouyrk (@cmagoyrk, Vice President at Oracle Cloud Infrastructure) about the architecture of the Oracle cloud, how it manages bare-metal VMs and containers, how Oracle integrates on-premises and cloud workloads, and how it integrates with PaaS and SaaS services.

            Show Links:

            Show Notes
            • Topic 1 - Welcome to the show. Can you tell us about your background prior to joining Oracle and how it helped you design the Oracle Cloud Infrastructure?
            • Topic 2 - There was a lot of buzz about Oracle redesigning their infrastructure cloud. Can you walk us through the architecture and how Oracle delivers bare-metal, VMs and containers? Are they separate environments, or all integrated?
            • Topic 3 - How does Oracle Cloud deliver networking - intra-cloud, from region-to-region, and integrated with on-premises applications?
            • Topic 4 - Oracle often talks about “same cloud architecture in both public and private”. Other public clouds are designed differently than most on-premises environments. Can you talk about what’s the same and what is typically different in the Oracle Cloud? What do customers have to adjust to in the Oracle Cloud?
            • Topic 5 - How do Oracle customers manage to integrate different Oracle Cloud services (e.g. containers, PaaS services, SaaS services, security, etc.)?
              Feedback?
              Posted by
              Labels: , , , , , , , ,

              Thursday, January 11, 2018

              The Cloudcast #329 - Tech Trade Shows in 2018

              Aaron and Brian attend a lot of technology events throughout the year. They give their suggestions to trade show organizers and attendees for getting the most of out 2018 events.

              Show Links:

              Show Notes
              • Topic 1 - What events do you enjoy attending? Not enjoy attending?
              • Topic 2 - What works for you at events?
              • Topic 3 - What could be improved or added?
              • Topic 4 - What should be eliminated, or re-imagined?
              • Topic 5 - Should we see more “independent” events combined (or parallel to) “vendor” events?
                Feedback?
                Posted by
                Labels: , , , , ,
                Subscribe to: Posts (Atom)