Sunday, February 27, 2022

The Kubernetes Developer Experience?

Kubernetes won the container wars and continues to grow in use across many industries. But how did something that was about Cloud-native Applications gain traction without a developer experience?

SHOW: 595

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:

 

HOW DID KUBERNETES WIN WHEN IT STARTED FROM BEHIND?

Listening to this week's SDT show, and remembering listening to SDT years ago, @cote comments about why Kubernetes "won" were always interesting. In essence it was late to market, was lacking in features vs. competitors (Mesos, Swarm, CF), and had a terrible user-experience...so how did it "win"? It all seems ass-backwards. 

HOW HAS KUBERNETES CONTINUED TO WIN, WITHOUT A DEVELOPER EXPERIENCE?

  1. Mesos, CF and Swarm were all single-vendor dominated projects, and many companies had concerns about another generation of vendor lock-in. This point is reasonably valid, but the companies that were using Mesos, CF and Swarm did all seem to love that technology.
  2. Mesos was primarily focused on big data workloads. For each new application-type, you needed to write (or use) another application-specific framework. So it was good at its niche, but couldn't easily be used for other types of apps. [Kubernetes eventually copied this model with CRDs].
  3. Swarm was the easiest to use, but it wasn't very good technology and didn't scale. So it got pigeon-holed for smaller projects.
  4. CF focused on Java/SpringBoot, which is a big Enterprise opportunity. but CF was super complicated to set up. And CF never really embraced containers, so companies were weary of if they were missing this big trend (Docker).
  5. Kubernetes comes along and  becomes the good-enough platform. It's not dominated by a single vendor. It natively supports Docker, it has some built-in usage patterns so it's easier than Mesos to add apps, it scales better than Swarm, and it can support Java/Spring or even legacy Java (lift-and-shift). And as Joe Beda says, you could use it natively or you could build some PaaS-y like features on top of it.

FEEDBACK?

Posted by

Wednesday, February 23, 2022

Data Observability

Kevin Hu (@kevinzenghu, Co-Founder | CEO at @Metaplane) talks about the concepts behind Data Observability and the unique challenges for Data Engineers.

SHOW: 594

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:


Topic 1 - Welcome to the show. Let’s talk about your background and what led you to start Metaplane.

Topic 2 - Let’s start by talking about the concept of what is a modern data engineer. What is this person doing, what are they responsible for, and who are their typical “customers” within a business. 

Topic 3 - Beyond just huge volumes of data and trying to make the data usable (formatting, ETL, storage access, etc.), what sort of problems do data engineers encounter? How much is typically “first-party data” and how much comes from external systems? 

Topic 4 - Let’s talk about Data Observability. First off, what is it?. And second, how is it different from the Observability that we’ve seen from Datadog or Honeycomb or Observe or many others? 

Topic 5 - What are the types of Data Observability problems that Metaplane is focused on solving for Data engineers? Are these usually done independently, or in collaboration with the application or business analyst teams?

Topic 6 - What are some of the immediate results (improvements) that companies see when adding Data Observability to their environments?


FEEDBACK?

Posted by
Labels: , ,

Sunday, February 20, 2022

DevOps, what's in a name?

DevOps is defined as, “Everything you do to overcome the friction created by silos … All the rest is plain engineering” Why is it so difficult to implement, but so easy to make up so many job titles surrounding it?

SHOW: 593

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:

 

IF EVERYTHING IS DEVOPS, THEN WHAT IS DEVOPS?

Patrick Dubois, one of the original creators of the DevOps concept, recently published a framework to think about all the job titles that have emerged out of the original DevOps concept. Is this a good progression, or is it just hiding the cultural and organizational complexities that DevOps is trying to address?

THE CHALLENGES OF TRYING TO ELIMINATE SILOED TECHNOLOGY FUNCTIONS

  1. It seems like it all began when AWS’ CTO Werner Vogels said, “You build it, you run it.” That was such a foreign concept in the IT world, and ever since, we’ve been trying to invent ways to apply existing organization paradigms to the byproduct of that concept. 
  2. Previous applications were rarely changed. Now applications (microservices, etc.) are designed from Day 1 to frequently change. Once again, the existing organization paradigms have to change. 
  3. And the public cloud introduces a level of financial transparency and complexity (e.g. here’s a bill every month) that is creating new roles (or newer concepts) that are trying to optimize that information. 
  4. Lots of companies say that they can’t support multiple operational models (hence why they try to retrofit old models into new names), but the reality is that they are always supporting multiple process or platform models. 
  5. Resume-driven-development has become a real thing, so it’s not unusual to see title-washing happen at various companies. It can also be valuable for recruiting as well, at least in the short-term. 
  6. I wish there was some better measurement model for progress along these DevOps paths, instead of just the tiering of the State of DevOps report. It’s too easy to get frustrated, when many companies are just striving to get to a mid-level of DevOps efficiency.

 

FEEDBACK?

Posted by

Wednesday, February 16, 2022

Cloud Cost Intelligence

Erik Peterson (@silvexis, Founder/CTO/CISO @CloudZeroInc) talks about how Cloud Cost Mgmt has matured, the importance of business context for cloud costs, and best practices for managing SaaS costs. 

SHOW: 592

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:

Topic 1 - Welcome to the show. You have a very interesting background. Tell us a little bit how you’ve gone from Nuclear watchdog to Banking to Product guy to Security guy and now focus on Cloud costs.

Topic 2 - Let’s start by talking about maturity levels of companies using the cloud. Engineering teams are fairly mature, but how mature are product teams, finance teams, marketing teams, etc, especially when it comes to understanding cloud costs in their world?

Topic 3 - Cloud Cost Mgmt used to be all about when to use on-demand vs. Reserved Instances. Have the systems evolved to be more contextual - (example) so marketing teams can run experiments and product teams can make smarter feature priority decisions?

Topic 4 - We’re seeing a lot of companies moving from selling software to delivering their capabilities as a SaaS offering. This is a big change in the economics of that business. How can those types of teams think about cloud costs?

Topic 5 - How much of Cost Mgmt is systems vs. people that understand the contexts of costs? What is your perspective on how much should be automated systems vs. human expertise to augment automated systems?

Topic 6 - What are some of the most common mistakes you see companies make around Cost Mgmt, or some tips you give to immediately help companies better spend on Cloud

FEEDBACK?

Posted by
Labels: , , , , , ,

Sunday, February 13, 2022

Lessons Learned from Shadow IT

Shadow IT and start-ups were the original users of public cloud, over a decade ago. But as public cloud has become a multi-billion dollar business, let’s explore how the role of Shadow IT has evolved. 

SHOW: 591

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

 

SHOW NOTES:

 

DOES SHADOW IT STILL EXIST IF PUBLIC CLOUD IS MAINSTREAM?

Shadow IT began as a way to be more productive in the office (server under the desk, WiFi in a conference room, etc.) and then it went to the cloud (SaaS, the IaaS/PaaS). But how did it evolve and what situations has it created now? 

WHAT DOES THE NEW DISTRIBUTED IT LOOK LIKE NOW?

  1. Everybody has the ability to get access to (almost) any technology, via open-source or public cloud, or freemium services.
  2. Everybody has the ability to learn something new (YouTube, ACloudGuru, Developer Evangelists, etc.
  3. IT organizations have less influence over company-wide architectures and strategies.
  4. IT still is often tasked with maintaining applications/security/compliance, even after another group deployed it. 
  5. IT leaders are asked to lead digital transformation projects, and typically aren’t staying in the same place for more than 2-3 years. How much of that time is spent coordinating, communicating, re-organizing around DevOps, DevSecOps, FinOps, AIOps, etc..
  6. There are hybrid applications, but they aren’t hybrid in the sense of consistently being deployed everywhere to manage vendor lock-in.
  7. There are many multi-cloud companies because IT no longer has a boundary. And the economics of cloud means that most applications won’t move once deployed (easier to turn off than to move).
  8. There are many, many pains-of-glass. 

FEEDBACK?

Posted by

Wednesday, February 9, 2022

Understanding & Managing Committed Cloud Spend

Tim Banks (@elchefe, Principal Cloud Economist @duckbillgroup) talks about how public cloud providers are offering committed spend programs, and ways to best use their programs to manage cloud spending. 

SHOW: 590

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:


Topic 1 - Welcome to the show. You’ve got a very interesting background of hands-on technical and being on the business side of things. Tell us a little bit about your background and what you focus on at Duckbill Group. 

Topic 2 - Let’s start by talking about this trend of long-term cloud contracts. Why are we seeing more and more announced, and who does it seem to benefit more (cloud provider or customer)?

Topic 2a - How do companies typically size these deals? Is it some percentage of current spend forecasted forward, or some aspirational goal, or something else? 

Topic 3 - When a company signs up for one of these long-term committed spend contracts, what are the mechanics of the contract? Is it just “all-you-can-eat” technology, or do they tend to include additional capabilities/services, etc?

Topic 4 - What have you found to be the behavior of companies that sign these contracts? Does it lead to more projects getting created, or more experimentation, or any other unintended consequences?

Topic 5 - At what point do companies start re-evaluating the contracts? What happens if they find themselves way below expected spending expectations?  

Topic 6 - Have you seen any new behaviors from the cloud providers once they sign a contract, whereas one group (or service) is pushing hard to capture a bigger portion of the contract? 

FEEDBACK?

Posted by
Labels: , , , , , ,

Sunday, February 6, 2022

The Parallel Lives of Kubernetes and Serverless

Our industry is fascinated with the origin stories of founders and the end of technology trends. In between, we seem fine with learning curves and technical debt.  

SHOW: 589

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:

 

WHEN ONE TECHNOLOGY ENDS, DOES ANOTHER TAKE ITS PLACE?

During a recent interview, I was asked “How close to complete is Kubernetes?”. We also saw a survey saying that less people are interested in Serverless (knowledge/training). Kubernetes is far from complete, and Serverless was supposed to be the next thing.   

WHY DO WE WORRY MORE ABOUT THE END OF A TECHNOLOGY AND LESS ABOUT TECHNICAL DEBT?

  1. Is the technology media only interested in origin stories and ending stories?
  2. Kubernetes is far from being complete, both as a project and the adjacencies around it. Why are we more interested in the end than in the middle (e.g. developer experience)?
  3. Serverless was supposed to replace Kubernetes - why have we lost interest in Serverless?
  4. Serverless didn’t have any “wars”, and it didn’t have any major startups or acquisitions (at least not yet)

FEEDBACK?

Posted by

Wednesday, February 2, 2022

Building Cloud Services from Software

Preeti Somal (@psomal, EVP Engineering @Hashicorp) talks about the evolution of Hashicorp Cloud Platform, creating cloud services from software projects, and the evolution of cloud SREs and SLAs.

SHOW: 588

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

CHECK OUT OUR NEW PODCAST - "CLOUDCAST BASICS"

SHOW SPONSORS:

SHOW NOTES:

Topic 1 - Welcome to the show. We’re always excited to have women leaders on the show. Tell us a little bit about your background and your focus areas at Hashicorp. Congrats on the UGA National Championship.

Topic 2 - Let’s talk about the Hashicorp Cloud Platform. We’ve covered Hashicorp software for years. How did HCP come about, and how has it changed how Hashicorp thinks about what you’re building (software vs. cloud services)?

Topic 3 - To create a cloud service out of an existing Hashicorp product (e.g. Terraform or Vault), does it just require creating an SRE team to run it, or do you have to rethink how the technologies are built and operated?

Topic 4 - How have Hashicorp customers adapted to consuming cloud services vs. running the software themselves? Sometimes the cloud services have to be more restrictive to deliver SLAs.

Topic 5 - Are you able to do anything unique in the cloud(s) that you couldn’t do as software products? Does multi-cloud provide any unique opportunities?

Topic 6 - Are you finding any new ways to interact with your communities now that Hashicorp is offering cloud services? 

FEEDBACK?

Posted by
Labels: , , , , ,
Subscribe to: Posts (Atom)