Please come over and collaborate with me there!
Joining us this week is Ian Rae, CEO and Founder CloudOps who recorded the podcast during the Google Next conference in 2018.
- 1 min 55 sec: Define Cloud from a CloudOps perspective
- Business Model and an Operations Model
- 3 min 59 sec: Update from Google Next 2018 event
- Google is the “Engineer’s Cloud”
- Google’s approach vs Amazon approach in feature design/release
- 9 min 55 sec: Early Amazon ~ no easy button
- Amazon educated the market as industry leader
- 12 min04 sec: What is the state of Hybrid? Do we need it?
- Complexity of systems leads to private, public as well as multiple cloud providers
- Open source enabled workloads to run on various clouds even if the cloud was not designed to support a type of workload
- Google’s strategy is around open source in the cloud
- 14 min 12 sec: IBM visibility in open source and cloud market
- Didn’t build cloud services (e.g. open a ticket to remap a VLAN)
- 16 min 40 sec: OpenStack tied to compete on service components
- Couldn’t compete without Product Managers to guide developers
- Missed last mile between technology and customer
- Didn’t want to take on the operational aspects of the customer
- 19 min 31 sec: Is innovation driven from listening to customers vs developers doing what they think is best?
- OpenStack is seen as legacy as customers look for Cloud Native Infrastructure
- OpenStack vs Kubernetes install time significance
- 22 min 44 sec: Google announcement of GKE for on-premises infrastructure
- Not really On-premise; more like Platform9 for OpenStack
- GKE solve end user experience and operational challenges to deliver it
- 26 min 07 sec: Edge IT replaces what is On-Premises IT
- Bullish on the future with Edge computing
- 27 min 27 sec: Who delivers control plane for edge?
- Recommends Open Source in control plan
- 28 min 29 sec: Current tech hides the infrastructure problems
- Someone still has to deal with the physical hardware
- 30 min 53 sec: Commercial driver for rapid Edge adoption
- 32 min 20 sec: CloudOps building software / next generation of BSS or OSS for telco
- Meet the needs of the cloud provider for flexibility in generating services with the ability to change the service backend provider
- Amazon is the new Win32
- 38 min 07 sec: Can customers install their own software? Will people buy software anymore?
- Compare payment models from Salesforce and Slack
- Google allowing customers to run their technology themselves of allow Google to manage it for you
- 40 min 43 sec: Wrap-Up
Podcast Guest: Ian Rae, CEO and Founder CloudOps
Ian Rae is the founder and CEO of CloudOps, a cloud computing consulting firm that provides multi-cloud solutions for software companies, enterprises and telecommunications providers. Ian is also the founder of cloud.ca, a Canadian cloud infrastructure as a service (IaaS) focused on data residency, privacy and security requirements. He is a partner at Year One Labs, a lean startup incubator, and is the founder of the Centre cloud.ca in Montreal. Prior to clouds, Ian was responsible for engineering at Coradiant, a leader in application performance management.
Joining us this week is Yves Boudreau from Ericsson for his 2nd Podcast appearance (1st Podcast) to talk about the new State of the Edge Report and the latest happenings in the Edge community.
- Edge as an accelerant not having to wait until Edge is built completely
- Opportunity Cost using Edge as is; no time to wait
- Be Specific when Requesting Services
- Internet and Networks are Not Unlimited Pipes
- Interesting Use Cases for Edge – Augmented Reality, Drone, Cars, Batteries
- Cost savings of where the data processing is done
- Open Source software communities at the Edge
Topic Time (Minutes.Seconds)
Intro 0.0 – 1.22
State of the Edge Report 1.22 – 5.22 (STE Podcast) (https://www.stateoftheedge.com/)
Accessible Edge Environments 5.22 – 10.50 (Bulgaria)
Opportunity Cost and Missing Killer App 10.50 – 12.04
Edge Infrastructure as Cloud Development Paradigm 12.04 – 14.29
Elasticity Issues b/w Cloud and Edge 14.29 – 21.45
Innovators Dilemma for Cloud & Telcom 21.35 – 23.10
Favorite Use Cases for Infrastructure Edge 23.10 – 28.55 (Hanger Podcast)
Data Location and Data Sovereignty 28.55 – 31.03
Cost for Processing Power in Edge Devices 31.03 – 34.49 (SWIM.AI Podcast)
Free Software/ Open Source in Edge 34.49 – 46.58
Wrap Up 46.58 – END
Podcast Guest: Yves Boudreau, VP Partnership and Ecosystem Strategy
Mr. Boudreau is a 20 year veteran of the Digital, Telecom and Cable TV industries. From modest beginnings of one of the first cable broadband ISPs in Canada to the fast paced technology hub of Silicon Valley, Yves joined ERICSSON in 2011 as Vice President of Technical Sales Support and most recently has accepted a position as the VP of Partnerships and Ecosystem Strategy for the ERICSSON Unified Delivery Network. Previously, Mr. Boudreau has worked in R&D, Systems Engineering & Business Development for companies such as Com21 Inc., ARRIS Group (Cable), Imagine Communication (Video Compression) and Verivue Inc. (CDN). Yves now resides in Atlanta, Georgia with his wife Josée and 3 children. Mr. Boudreau completed his undergraduate studies in Commerce @ Laurentian University and graduate studies in Information Technology Management @ Athabasca University. Yves currently also serves on the Board of Director of the Streaming Video Alliance (www.streamingvideoalliance.org)
Note: OpenStack voting is limited to community members – if you registered by the deadline, you will receive your unique ballot by email. You have 8 votes to distribute as you see fit.
I believe open infrastructure software is essential for our IT future.
Open source has been a critical platform for innovation and creating commercial value for our entire industry; however, we have to deliberately foster communities for open source activities that connect creators, users and sponsors. OpenStack has built exactly that for people interested in infrastructure and that is why I am excited to run for the Foundation Board again.
OpenStack is at a critical juncture in transitioning from a code focus to a community focus.
We must allow the OpenStack code to consolidate around a simple mission while the community explores adjacent spaces. It will be a confusing and challenging transition because we’ll have to create new spaces that leave part of the code behind – what we’d call the Innovator’s Dilemma inside of a single company. And, I don’t think OpenStack has a lot of time to figure this out.
That change requires both strong and collaborative leadership by people who know the community but are not too immersed in the code.
I am seeking community support for my return to the OpenStack Foundation Board. In the two years since I was on the board, I’ve worked in the Kubernetes community to support operators. While on the board, I fought hard to deliver testable interoperability (DefCore) and against expanding the project focus (Big Tent). As a start-up and open source founder, I bring a critical commercial balance to a community that is too easily dominated by large vendor interests.
Re-elected or not, I’m a committed member of the OpenStack community who is enthusiastically supporting the new initiatives by the Foundation. I believe strongly that our industry needs to sponsor and support open infrastructure. I also believe that dominate place for OpenStack IaaS code has changed and we also need to focus those efforts to be highly collaborative.
OpenStack cannot keep starting with “use our code” – we have to start with “let’s understand the challenges.” That’s how we’ll keep building an strong open infrastructure community.
If these ideas resonate with you, then please consider supporting me for the OpenStack board. If they don’t, please vote anyway! There are great candidates on the ballot again and voting supports the community.
Welcome to the weekly post of the RackN blog recap of all things Digital Rebar, RackN, Edge Computing, and DevOps. If you have any ideas for this recap or would like to include content please contact us at email@example.com or tweet RackN (@rackngo)
Items of the Week
- Hot IoT tech trends for 2018 : Network World
Edge computing, in the context of IoT, is the idea that you can actually do some of the computational work required by a system close to the endpoints instead of in a cloud or a data center. The intent is to minimize latency, which, according to Renaud, means that it’s going to be a hot trend in certain kinds of industrial IoT application.
Solution providers that have been hit hard by a data center hardware retreat are finding sales and profit growth by living on the edge—the network edge, that is.
- Five DevOps trends to watch out for : RCR Wireless News
DevOps — a term used to refer to the integration of software developers and operations teams — continues to spread like wildfire throughout the open networking ecosystem. The main idea behind DevOps is that by breaking down barriers between these two departments, market applications can be delivered faster with lower costs and better quality. Nevertheless, for all the advantages attached to DevOps, it is still a budding concept since it is primarily concerned with re-aligning the workforce with a variety of tools. The following, therefore, is a list of DevOps trends to keep an eye out for.
Our architectural plans for Digital Rebar are beyond big – they are for massive distributed scale. Not up, but out. We are designing for the case where we have common automation content packages distributed over 100,000 stand-alone sites (think 5G cell towers) that are not synchronously managed. In that case, there will be version drift between the endpoints and content. For example, we may need to patch an installation script quickly over a whole fleet but want to upgrade the endpoints more slowly.
- Digital Rebar Community Online Meetup – December 5th at 11:00am PST
Prior Meetup on November 21st Notes
Yesterday, AWS confirmed that it actually uses physical servers to run its cloud infrastructure and, gasp, no one was surprised. The actual news about the i3.metal instances by AWS Chief Evangelist Jeff Barr shows that bare metal is being treated as just another AMI managed instance type (see also Geekwire, Techcrunch, Venture Beat). For AWS users, there’s no drama here because it’s an incremental add to processes they are already know well.
We are actively looking for feedback from customers and technologists before general availability of both RackN and the Terraform plug-in. It takes just a few minutes to get started and we offer direct engineering engagement on our community slack channel. Get started now by providing your email on our registration pagey so we can provide you all the necessary links.
L8ist Sh9y Podcast
Founder and Chief Research Advisor, Infrastructure, Application Platforms and DevOps
- KubeCon + CloudNativeCon : Dec 6 – 8 in Austin, TX
Event plans for the RackN and Digital Rebar team include 2 sessions and the RackN booth. We look forward to seeing you in Austin.
The RackN team is preparing for a series of upcoming events where they are speaking or just attending. If you are interested in meeting with them at these events please email firstname.lastname@example.org
- Gartner IT Infrastructure, Operations Management and Data Center – Dec 4 – 7 Event Link
- CloudNativeCon + KubeCon – Dec 8 : Zero Configuration Pattern of Kubernetes on Bare Metal
I love great conversations about technology – especially ones where the answer is not very neatly settled into winners and losers (which is ALL of them in IT). I’m excited that RackN has (re)launched the L8ist Sh9y (aka Latest Shiny) podcast around this exact theme.
Please check out the deep and thoughtful discussion I just had with Mark Thiele (notes) of Apcera where we covered Mark’s thought on why public cloud will be under 20% of IT and culture issues head on.
Spoiler: we have David Linthicum coming next, SO SUBSCRIBE.
We feel there’s still room for deep discussions specifically around automated IT Operations in cloud, data center and edge; consequently, we’re branching out to start including deep interviews in addition to our initial stable of IT Ops deep technical topics like Terraform, Edge Computing, GartnerSYM review, Kubernetes and, of course, our own Digital Rebar.
I’m investing in these Site Reliability Engineering (SRE) discussions because I believe operations (and by extension DevOps) is facing a significant challenge in keeping up with development tooling. The links below have been getting a lot of interest on twitter and driving some good discussion.
Addressing this Ops debt is our primary mission at my company, RackN: we believe that integrated system level tooling is required. We also believe that new tools should not disrupt environments so we work very hard to adapt to requirements of individual sites.
SRE is urgent because it provides a pragmatic path and rationale for investment.
Even if you don’t agree with Google’s term or all their practices, I think fundamental concepts of system thinking, status/pay, automation investment and developer collaboration are essential. It should come as no surprise that these are all Lean/DevOps concepts; however, SRE has the pragmatic side of being a job function.
Here are some recent relevant discussions I’ve been having about SREs with links to both the audio and my text show notes.
- Cloud Cast about SRE concepts and decomposing Ops
- Datanauts deep dive about SRE based on the “DevOps vs SRE” talk from DevOpsDays Austin (original post)
- Charity Majors and I debate the SRE name and pay equity for Ops.
- Further Reading Podcasts
Of course, RackN is also doing a WEEKLY SRE update that captures general interest items. Check that out and subscribe.
TL;DR: infrastructure operations is hard and we need to do a lot more to make these systems widely accessible, easy to sustain and lower risk. We’re discussing these topics on twitter…please join in. Themes include “do we really have consensus and will to act” and “already a solved problem” and “this hurts OpenStack in the end.”
I am always looking for ways to explain (and solve!) the challenges that we face in IT operations and open infrastructure. I’ve been writing a lot about my concern that data center automation is not keeping pace and causing technical debt. That concern led to my recent SRE blogging for RackN.
It’s essential to solve these problems in an open way so that we can work together as a community of operators.
It feels like developers are quick to rally around open platforms and tools while operators tend to be tightly coupled to vendor solutions because operational work is tightly coupled to infrastructure. From that perspective, I’m been very involved in OpenStack and Kubernetes open source infrastructure platforms because I believe the create communities where we can work together.
- How is OpenStack so dead AND yet so very alive to SREs?
- OpenStack’s Big Pivot: our suggestion to drop everything and focus on being a Kubernetes VM management workload
Of course, I do have a vested interest here. Our open underlay automation platform, Digital Rebar, was designed to address a missing layer of physical and hybrid automation under both of these projects. We want to help accelerate these technologies by helping deliver shared best practices via software. The stack is additive – let’s build it together.
I’m very interested in hearing from you about these ideas here or in the context of the individual posts. Thanks!
TL;DR: Sometimes paradigm changes demand a rapid response and I believe unifying OpenStack services under Kubernetes has become an such an urgent priority that we must freeze all other work until this effort has been completed.
See Also Rob’s VMblog.com post How is OpenStack so dead AND yet so very alive
By design, OpenStack chose to be unopinionated about operations.
That made sense for a multi-vendor project that was deeply integrated with the physical infrastructure and virtualization technologies. The cost of that decision has been high for everyone because we did not converge to shared practices that would drive ease of operations, upgrade or tuning. We ended up with waves of vendors vying to have the the fastest, simplest and openest version.
Tragically, install became an area of competition instead an area of collaboration.
Containers and microservice architecture (as required for Kubernetes and other container schedulers) is providing an opportunity to correct this course. The community is already moving towards containerized services with significant interest in using Kubernetes as the underlay manager for those services. I’ve laid out the arguments for and challenges ahead of this approach in other places.
These technical challenges involve tuning the services for cloud native configuration and immutable designs. They include making sure the project configurations can be injected into containers securely and the infra-service communication can handle container life-cycles. Adjacent concerns like networking and storage also have to be considered. These are all solvable problems that can be more quickly resolved if the community acts together to target just one open underlay.
The critical fact is that the changes are manageable and unifying the solution makes the project stronger.
Using Kubernetes for OpenStack service management does not eliminate or even solve the challenges of deep integration. OpenStack already has abstractions that manage vendor heterogeneity and those abstractions are a key value for the project. Kubernetes solves a different problem: it manages the application services that run OpenStack with a proven, understood pattern. By adopting this pattern fully, we finally give operators consistent, shared and open upgrade, availability and management tooling.
Having a shared, open operational model would help drive OpenStack faster.
There is a risk to this approach: driving Kubernetes as the underlay for OpenStack will force OpenStack services into a more narrow scope as an infrastructure service (aka IaaS). This is a good thing in my opinion. We need multiple abstractions when we build effective IT systems.
The idea that we can build a universal single abstraction for all uses is a dangerous distraction; instead; we need to build platform layers collaborativity.
While initially resisting, I have become enthusiatic about this approach. RackN has been working hard on the upgradable & highly available Kubernetes on Metal prerequisite. We’ve also created prototypes of the fully integrated stack. We believe strongly that this work should be done as a community effort and not within a distro.
My call for a Kubernetes underlay pivot embraces that collaborative approach. If we can keep these platforms focused on their core value then we can build bridges between what we have and our next innovation. What do you think? Is this a good approach? Contact us if you’d like to work together on making this happen.
See Also Rob’s VMblog.com post How is OpenStack so dead AND yet so very alive to SREs?
Author’s call to action: if you think you already know this is a problem, then why do we keep reliving it? We’re doing our part open with Digital Rebar and we need more help to secure infrastructure using foundational automation.
There’s a frustrating cyberattack driven security awareness cycle in IT Operations. Exploits and vulnerabilities are neither new nor unexpected; however, there is a new element taking shape that should raise additional alarm.
Cyberattacks are increasingly profit generating and automated.
The fundamental fact of the latest attacks is that patches were available. The extensive impact we are seeing is caused by IT Operations that relies on end-of-life components and cannot absorb incremental changes. These practices are based on dangerous obsolete assumptions about perimeter defense and long delivery cycles.
It’s not just new products using CI/CD pipelines and dynamic delivery: we must retrofit all IT infrastructure to be constantly refreshed.
We simply cannot wait because the cybersecurity challenges are accelerating. What’s changed in the industry? There is a combination of factors driving these trends:
- Profit motive – attacks are not simply about getting information, they are profit centers made simpler with hard to trace cryptocurrency.
- Shortening windows – we’re doing better at finding, publishing and fixing issues than ever in the open. That cycle assumes that downstream users are also applying the fixes quickly. Without downstream adoption, the process fails to realize key benefit.
- Automation and machine learning – the attackers are using more and more sophisticated automation to find and exploit vulnerabilities. Expect them to use machine learning to make it even more effective.
- No perimeter – our highly interconnected and mobile IT environments eliminate the illusion of a perimeter defense. This not just a networking statement: our code bases and service catalogs are built from many outside sources that often have deep access.
- Expanding surface area – finally, we’re embedding and connected more devices every second into our infrastructure. Costs are decreasing while capability increases. There’s no turning back from that, we we should expect an ongoing list of vulnerabilities.
No company has all the answers for cybersecurity; however, it’s clear that we cannot solve this cybersecurity at the perimeter and allowing the interior to remain static.
The only workable IT posture starts with a continuously deployed and updated foundation.
Companies typically skip this work because it’s very difficult to automate in a cross-infrastructure and reliable way. I’ve been working in this space for nearly two decades and we’re just delivering deep automation that can be applied in generalized ways as part of larger processes. The good news is that means that we can finally start discussing real shared industry best practices.
Thankfully, with shared practices and tooling, we can get ahead of the attackers.
RackN focuses exclusively on addressing infrastructure automation in an open way. We are solving this problem from the data center foundations upward. That allows us to establish security practice that is both completely trusted and constantly refreshed. It’s definitely not the only thing companies need to do, but that foundation and posture helps drive a better defense.
I don’t pretend to have complete answers to the cyberattacks we are seeing, but I hope they inspire us to more security discipline. We are on the cusp of a new wave of automated and fast exploits.
Let us know if you are interested in working with RackN to build a more dynamic infrastructure.