The Cloud Pod

Latest Episodes:

171: AWS Snowcones in Space Jun 30, 2022

Show notes

On The Cloud Pod this week, Peter finally returns with some beer-based bets about Amazon extending its TLS deadline. Plus: Terraform drift detection for managing infrastructure, chilling tales of Amazon’s CodeWhisperer ML advances, and Anthos on-premise options finally arrive for your platform of choice. Plus the cloud talks about AWS SNOWCONES in SPACE!!!!!!

A big thanks to this week’s sponsor, Foghorn Consulting, which provides full-stack cloud solutions with a focus on strategy, planning and execution for enterprises seeking to take advantage of the transformative capabilities of AWS, Google Cloud and Azure.

This week’s highlights

Terraform Cloud finally adds drift detection to help manage infrastructure, now generally available after its 2020 preview.
Amazon’s crazy “ML-powered coding companion,” CodeWhisperer, is here for our jobs.
Google expands its Distributed Cloud platform with Anthos on-premises options.

Top Quotes

“I’m surprised it’s taken so long. Because I mean, the reality is if you’re in a plan, and the plan doesn’t require any changes, then there’s been no drift. So what was the obstacle in delivering this as a feature sooner?”
“Not only they’re training their own machine learning models, but they’re also generating code. Not concerned at all.”

General News: Drifting in the Right Direction

While everyone’s been a little afraid to pull the trigger, HashiCorp announced drift detection in Terraform cloud, which is in a public beta. Pretty exciting!
HashiCorp also announced the launch and free public beta of HCP Boundary, but what’s their long-term vision?

AWS: Whispering Sweet Somethings to the Machine

SageMaker Ground Truth now supports synthetic data generation, promising to reduce time and training costs for model operations. Getting enough data to actually train a model could be hard… (fake it til you make it?)
Your new “ML-powered coding companion” CodeWhisperer now writes code for you. We’ve joked about it before, but Alexa really is one step away from upskilling to coding.
Peter’s betting two beers at his local pub on Amazon extending the deadline on this one: TLS 1.2 is to become the minimum TLS protocol level for all AWS API endpoints. There’s currently just under a year to get yourself sorted. Good luck!
Apparently, even space has (AWS) Snowcones: Amazon sends one to the International Space Station
As EKS improves control plane scaling and update speed by up to 4x, get ready for a lot of step function workload.
Imagine waiting 10 years for private IP VPNs… well, we did, and

170: The Cloud Pod Is Also Intentionally Paranoid Jun 23, 2022

Show notes

On The Cloud Pod this week, the team discusses Jonathan’s penance for his failures. Plus: Microsoft makes moves on non-competes, NDAs, salary disclosures, and a civil rights audit; AWS modernizes mainframe applications for cloud deployment; and AWS CEO Adam Selipsky chooses to be intentionally paranoid.

This week’s highlights

The Balmer era is officially dead: Microsoft curbs non-competes, drops NDAs from worker settlements, disclose salary ranges, and even launches a civil rights audit.
AWS launches their new modernization service for mainframe applications, now deployable in fully managed AWS runtime environments.
AWS CEO Adam Selipsky “choose[s] to be intentionally paranoid,” as he leads the company through turbulence.

Top Quotes

“We’ve talked about how garbage those [noncompetes] are, the problems they’ve had with them, executives leaving, Amazon going to Microsoft, then getting sued and all the mess of that. So I’m super glad they’re finally starting to see a tide swell change in technology where that’s no longer a thing.”
“I always felt like Amazon was going to just create a mainframe as a service offering — buy a bunch of IBM mainframes that they sell out to you — because that’s been a model of mainframe for a long time: CPU slicing, rentals and that kind of thing. But it seems like now they’re going to go down this other path where the answer is [that] you convert to a more modern architecture, which is interesting.”

General News: It’s a New Era

The times they are a-changin’, as Microsoft revises its position on non-competes, NDAs, and salary range disclosure, while launching a civil rights audit. Take that, Amazon!
Target CIO Mike McNamara jumps away from AWS with a scaled move toward multicloud architecture. Target allegedly has 4,000 engineers, which seems like a lot.
Archera vents via Venturebeat about the unmanageability of cloud costs, calling for standardized billing. While it might be helpful and even valuable, this seems a road too far traveled.

AWS: Modernized Mainframes and Intentional Paranoia

You can now take advantage of AWS’ new modernization service for mainframe applications, deployable in fully managed AWS runtime environments.
There are some nice enhancements for MGN, including DR configuration and Linux to Rocky Linux and SUSE Linux Subscription conversions.
AWS CEO Adam Selipsky admits, “I choose to be intentionally paranoid,” as he leads the company into a turbulent world.
A nice feature so

169: The CloudPod bounces back with Elastic Disaster Recovery Jun 08, 2022

Show notes

On The Cloud Pod this week, half the team whizzes through the news in record time. Plus: AWS Elastic Disaster Recovery, Google Distributed Cloud adds AI, ML and Database Solutions, and there’s another win for NetApp with Azure VMware Solution.

This week’s highlights

AWS Elastic Disaster Recovery now supports up to 300 staging and target accounts, which seems like a small number for some enterprises with thousands.
With the power of Anthos, Google Distributed Cloud adds AI, ML and Database Solutions — continuing the trend of service monetization regardless of host location.
Another win for NetApp, the home of choice for Azure VMware solutions optimization.

Top Quotes

“If you’re really doing auto scaling [and] traditional cloud native, you don’t use the service because you’ve already built it into your app. So this is for legacy IT operations like SAP, Oracle, and others. Three hundred or 3,000 covers small and medium business, but large enterprise has way more than that.”
“When Anthos first was announced, and Outpost for AWS, we talked about how likely it was that more and more cloud-native services were going to be made available anywhere, on any cloud, in any data center. It’s definitely a pattern of monetizing the services regardless of where they’re hosted.”

AWS: Bouncing Back From Disaster

Amazon EMR Serverless is now generally available, a cool feature running big data applications (and Outpost too). But it’s interesting that it’s been branded “serverless” when it’s clearly a managed service.
Elastic Disaster Recovery now supports 300 staging and target accounts, but we can’t help wondering how this helps the largest enterprises.
Step Functions launches a workflow-based interactive application workshop, and it looks like a golden age for developer experience is close at hand.
Amazon Route 53 announces IP-based routing for DNS queries, which is going to make things complicated. So preoccupied with whether or not they could integrate, they didn’t stop to think if they should.

GCP: Complexity on Top of Complexity

Google Chronicle offers context-aware detections, alert prioritization and risk scoring for its Security Operations. But wouldn’t you want to protect everybody from everything?
A boon for customer choice and flexibility: Google Distributed Cloud adds AI, ML and database solutions. On prem, running Kubernetes and Anthos? Justin loves this.
Yeehaw! Time to grab that 10-gallon hat and run you

168: The Cloud Pod Celebrates GCP Madrid Region With Sangria Jun 03, 2022

Show notes

On The Cloud Pod this week, the team discusses the new Madrid region’s midday siesta shutdown. Plus: Broadcom acquires VMWare for $61 billion, Azure gets paradigmatic with 5G, and you can now take the 2022 Google-DORA DevOps survey.

This week’s highlights

Broadcom acquires VMWare for $61 billion, in one of the largest-ever acquisitions.
Google Cloud and DORA team up to bring us the 2022 Accelerate State of DevOps Survey.
Azure calls 5G a “paradigm,” but is it just hype?

Top Quotes

“This is an interesting reverse on the large cloud providers getting into the silicon business, which makes sense to me — that they want to control their supply chain and optimize. … Is Broadcom going to start becoming like a cloud provider? That’s interesting. I wouldn’t suspect that.”
“What [is Azure] trying to do? Are they trying to sell us on [5G]? Are they trying to change the way we develop? Because we’re just going to waste our time developing stuff that requires some of these things, and then the infrastructure is not going to be there to support it.”

General News: Diversifying the Portfolio

In one of the largest acquisitions ever (just shy of Dell’s EMC takeover at $67 billion and Microsoft’s Blizzard acquisition at $69 billion), Broadcom acquires VMware for $61 billion. This could have big implications for enterprise.

AWS: Need for Speed

If you need a lot of disk space to log transactions, you’re in luck: Amazon EC2 M6id and C6id instances buff up their storage by up to 7.6TB.
Ryan’s usually doing whatever he can to avoid this, but if you need Elastic Volumes and Fast Snapshot Restore (FSR) support for io2 Block Express, you’ve now got it.

GCP: the State of DevOps in 2022

Why do IT leaders choose Google Cloud certification for their teams? In case you were wondering, here’s a puff piece with the answer.
If you need to change streams with Cloud Spanner, you can now do so. A cool feature, but it does need to be by email (there’s no homing pigeon option… yet).
If you want to learn a whole bunch of irrelevant HPC jargon, this is the blog post for you.
You can now take the 2022 Accelerate State of DevOps Survey, launched by Google and DORA.

167: The Cloud Pod Gets Sucked In by the Graviton3 May 25, 2022

Show notes

On The Cloud Pod this week, the team talks tactics for infiltrating the new Google Cloud center in Ohio. Plus: AWS goes sci-fi with the new Graviton3 processors, the new GKE cost estimator calculates the value of your soul, and Microsoft builds the metaverse.

This week’s highlights

AWS fires up the Graviton3 processors for some big energy savings.
Google develops the new GKE cost estimator for people who aren’t curious about cost.
Microsoft Build comes out of nowhere to deliver awesome, scary AI-driven tools with much mention of metaverse (yuck).

Top Quotes

“This feature isn’t developed for you because you’re curious about the cost. This is developed specifically for the people who are not curious about the cost. It’s a big red number. When they’re doing the deployment, it’s like, oh, I should probably not do that.”
“I cannot wait for the robot overlords to completely school me at code. This is gonna be hilarious… and frightening.”

General News: HashiCorp Extends Its Reach

Ryan is slightly embarrassed by how much he’s excited about the new HCL Extension for Visual Studio Code 0.1 announcement.

AWS: Abiding by the Laws of Graviton3

Storage company NetApp continues to buck industry trends with Backup and FSx support for ONTAP. Don’t forget to check out the TCP Talks interview with Anthony Lye, Executive VP and General Manager of NetApp.
New AWS-designed Graviton3 Processors power Amazon EC2 C7g Instances, now generally available.
Control Tower now supports concurrent operations for preventive guardrails. Awesome if you’re just starting, tougher if you’ve been at it for a while.
If you’ve been waiting for Kendra to give you something you actually cared about in dev, here you go: Jira connector enables document search on Jira repository.
Great news: Incident Manager expands support for runbook automation. We love announcements like these.
Ryan now has even less excuse for not trying Resilience Hub, after it adds support for Terraform, Amazon ECS and more.
Once again, AWS admits that multicloud is a real thing, with

166: The Cloud Pod Eagerly Awaits the Microsoft Pay Increase May 18, 2022

Show notes

On The Cloud Pod this week, the team struggles with scheduling to get everyone in the same room for just one week. Plus, Microsoft increases pay for talent retention while changing licensing for European Cloud Providers, Google Cloud introduces AlloyDB for PostgreSQL, and AWS announces EC2 support for NitroTPM.

This week’s highlights

Big changes are afoot with Microsoft on both pay and European licensing fronts.
A very busy Google finds time to release AlloyDB for PostgreSQL.
NitroTPM gets Amazon EC2 support.

Top Quotes

“I hope that it’s the exact opposite of TK and Google Cloud — that they’re really focused on the values and the culture and providing meaningful work. Especially during the last year in the pandemic, a lot of people have realized there’s a lot of different priorities; that money is good — it doesn’t buy happiness, but it buys a lot of things that can make me happy — but it’s getting that fulfillment, and enrichment is also super important. Not just a slog.”
“The problem is they’re not building power plants fast enough to support all of the power demand they have in this country. So there’s a possibility that these cloud providers may get pushback on building data centers in the region, which can have a huge detrimental impact. So keep an eye on that.”

AWS: Some Dynamite Announcements

AWS teams up with IBM in a SaaS-based partnership. Interesting that it’s IBM, but money talks, and there’s no better time to do it.
EC2 now supports NitroTPM and UEFI Secure Boot, which is an interesting pivot for the security-minded.
Open source supply chain security gets a nice big $10 million investment from AWS.
If you need the functionality, you’ve got some nice EKS Anywhere curated software packages to choose from, which are now in public preview.
CloudWatch improves the console experience, which no one really wants. There’s a lot more Amazon can be doing.

GCP: Busy Little Bees

AlloyDB for PostgreSQL promises freedom from expensive legacy databases. Here’s to hoping it works.

164: The Cloud Pod SWIFT-ly Moves Its Money to Google Cloud May 11, 2022

Show notes

On The Cloud Pod this week, Peter’s been suspended without pay for two weeks for not filing his vacation requests in triplicate. Plus it’s earnings season once again, there’s a major Google and SWIFT collaboration afoot, and MSK Serverless is now generally available, making Kafka management fairly hassle-free.

This week’s highlights

Earnings season is upon us once again, with billions earned and lost. Who are the winners?
MSK Serverless is now generally available as a boon for Kafka management.
Google and SWIFT uproot the financial world in announcing a huge cloud-based collaboration.

Top Quotes

“It’s hard to call a 32% increase for Azure earnings a slowdown, but it is definitely slower than what they saw in 2021 and the boom of the pandemic. But the overall trend is everyone’s gonna keep adopting cloud hyperscalers to host their infrastructure.”
“The important thing about this is that it’s signaling a change in compliance controls; all these financial organizations with very traditionally physical hardware in Iraq in the data center [had] no way to move to the cloud. So whether it’s through advocacy or proof of process, being able to virtualize all these things is going to be huge and will open up a massive market for new customers.”

General News: Earnings Are In, and It’s Looking… Good?

Imagine earning $116.4 billion and then still losing money. But fear not after such a rough quarter, Amazon: AWS revenue is here to save the day at 37%.
Meanwhile, Google revenue increased slightly below expectations, and GCP is still losing money — but $43 million less than last year.
Finally, Microsoft has Azure to thank for its 32% growth.

AWS: A Truly Kafkaesque Affair

MSK Serverless is now generally available, offering a reduction in the overhead of managing Kafka.
Amazon EC2 instances get some storage-optimizing icy processing power. (You just know there’s still a whole team of DBAs that doesn’t think this is good enough.)
Last on the AWS front: There are new management features for EC2 key pairs. We’re ecstatic!

GCP: Last Chance to Register for the Google Cloud Security Summit

GCP offers some CISO perspectives on security updates, as well as a reminder to register for the upcoming summit.
No-code solutions provide some nightmare fuel, as

165: The Cloud Pod Angry That Amazon Describes Step Functions as Low Code May 11, 2022

Show notes

On The Cloud Pod this week, the team discusses wholesome local Oakland toast for breakfast. Plus: Hybrid infrastructure is unsustainable, the AWS Proton template library expands, and Amazon angers the team by describing Step Functions as “low-code.”

This week’s highlights

Against the trend of popular opinion, it turns out that hybrid infrastructure is a bad idea in the long term, with a few significant drawbacks.
The AWS Proton template library just got bigger, so now people can find something else to complain about.
Amazon annoyingly describes Step Functions as low-code, which is definitely not true.

Top Quotes

“Proton was only developed as an answer for, how should we deploy onto Amazon? It’s setting yourself up just so someone can armchair-quarterback and poke holes in it. Now they’re saying, well, how would you do this? [Answer:] You have the templates. And then they’re gonna be like, the templates are cool, except it doesn’t meet my pretty edge case, so they’ll complain about that. We’ll see templates for the templates next.”
“I just love the assumption that you could low-code a solution with Step Functions, just because I’ve created many a step function and state machine flow. And all it is is coding and then figuring out why the code isn’t doing what I want — because I’m not passing things correctly between the different functions. The ability for someone who can’t write code to be able to to accomplish anything is a little far fetched.”

General News: Don’t Plan on Hybrid for Long…

In the cloud court of public opinion, dissent is infrequent. Yet here’s Michael Bathon of Rimini Street claiming that hybrid is actually bad in the long-term.

AWS: What Is Low-Code, Anyway?

The AWS Proton template library expands — as does people’s list of things to complain about.
Amazon very irritatingly calls Step Functions low-code, with new workflow observability features.
Can the annoying customer with the single use case please stand up? Amazon RDS for PostgreSQL now supports a lot more read replicas. Driven by the business side, perhaps?

GCP: Something’s Got To Give With BigQuery

Cloud TPU VMs are now generally available, with faster speeds and lower costs for training.
BigQuery BI Engine now supports more tools and custom applications. All we heard is that the analysts want to learn BigQuery, so they made it work for them.
It’s one thing to provide a good service and another thing to develop an open source tool that

162: The Cloud Pod Catches a Fleeting Glimpse of Google Cloud Optimization Apr 29, 2022

Show notes

On The Cloud Pod this week, the team rediscovers who Ryan is after an eternity (a secret agent). Plus AWS Fargate now delivers faster scaling of applications; new features for Oracle Support Rewards; and Google Cloud Optimization AI: Cloud Fleet Routing API from GCP.

This week’s highlights

Witness the magic of AWS Fargate scaling of applications — harder, faster, better, stronger.
Ooooh! Unveiling brand new shiny features for Oracle Support Rewards.
Better planning with more routes: GCP unleashes Optimization AI, API for Cloud Fleet Routing (CFR).

Top Quotes

“Because of that Fargate-specific limitation, [with] the first three services you’re concurrently updating, you’ll actually get a much faster rate through ECS test launches, but that fourth service will be slower. At that point, if the math works out where you’re better off hosting it on EC2 … it’s a lot more complex. I’ve worked with a lot of teams on trying to get ECS services to scale faster, and usually I look at them a little skeptically — do you really need this fast?”
“In terms of looking at lists of interview questions from Google algorithm questions and the traveling salesman problem and optimizing journeys through multiple locations, multiple cities, everything else, it’s a really hard problem. It only gets exponentially more difficult. And then the more efficient you are with that, the more it costs the environment, the more it costs in time or it costs money. So yeah, it’s actually a worthy problem to solve.”

General News: Microsoft Feels the Heat

We’re feeling the pain of Microsoft’s licensing, as its tactics to win the cloud battle lead to new antitrust scrutiny.

AWS: A Very Fargate Indeed

NetApp’s ONTAP, so line up your glasses for a very fine update indeed. Check out the podcast where we interviewed their very own Anthony Lye. #ShamelessPodcastSalesmanship
AWS Fargate now delivers faster scaling of applications, and you can see it in action with ECS. Understand token buckets and how AWS uses them, and if you need a hero, Vlad Ioenscu is here.
Microsoft Active Directory geeks rejoice: a favored topic of the masses with configurable synchronization launched via Single Sign-On.
The Log4j saga simply won’t die: Apache hotpatch issues get

163: The Cloud Pod Pushes the Azure Red Button Apr 27, 2022

Show notes

On The Cloud Pod this week, the team establishes that Justin may be immune to COVID. Plus all the latest from the AWS Summit, Azure Red Button team up on DDOS defense, and engines are revving in the great VMware showdown.

This week’s highlights

The AWS San Francisco Summit kicks off with a ton of new generally available stuff, but not-so-impressive attendance (looking at you, COVID).
Microsoft and Red Button buddy up on DDOS defense testing initiative.
AWS, Google and Oracle rev their engines for the VMware top spot.

Top Quotes

“Really shows you the power of partnership … There’s finally some easy button for testing these things. Because you always dream: Maybe I could create my own DDoS situation, which seemingly I do occasionally by accident, but intentionally would be nice this time.”
“I don’t necessarily trust their math, but assuming that it’s reasonably correct, it seems like a good market for Oracle to go after if you’re gonna try to compete with those three platforms — I don’t see a ton of people moving straight to the cloud on VMware. But that’s a pretty compelling argument and potentially a way of getting VMware customers to the cloud quicker: let’s just do it now if we don’t have to get off of VMware.”

General News: Great Expectations

Gartner anticipates big growth (20.4%) in public cloud spending for 2022!

AWS: Everything Generally Available

Finally, you can use IAM to control access to a resource based on the account, OU or organization that contains the resource — just how it used to be, and makes a whole lot more sense.
You might be excited for the confusingly named Amazon CloudWatch for Ray — if you can work out what it is (we couldn’t). Something to do with machine learning?
One for the data scientists: Announcing the Amazon SageMaker Serverless Inference, which should prove a boon for infrastructure management.
Now the guru can tell you your code sucks, too: Introducing the power of operational issue automatic detection in Lambda Functions with Amazon DevOps Guru for Serverless.
IoT TwinMaker is now generally available, and while your host doesn’t understand, luckily Ryan is on hand to talk about its uses.
AWS Amplify Studio is also now

Our TOPPODCAST Picks

Follow Us

Stay Connected

This week’s highlights

General News: Drifting in the Right Direction

AWS: Whispering Sweet Somethings to the Machine

This week’s highlights

General News: It’s a New Era

AWS: Modernized Mainframes and Intentional Paranoia

This week’s highlights

AWS: Bouncing Back From Disaster

GCP: Complexity on Top of Complexity

This week’s highlights

General News: Diversifying the Portfolio

AWS: Need for Speed

GCP: the State of DevOps in 2022

This week’s highlights

General News: HashiCorp Extends Its Reach

AWS: Abiding by the Laws of Graviton3

This week’s highlights

AWS: Some Dynamite Announcements

GCP: Busy Little Bees

This week’s highlights

General News: Earnings Are In, and It’s Looking… Good?

AWS: A Truly Kafkaesque Affair

GCP: Last Chance to Register for the Google Cloud Security Summit

This week’s highlights

General News: Don’t Plan on Hybrid for Long…

AWS: What Is Low-Code, Anyway?

GCP: Something’s Got To Give With BigQuery

This week’s highlights

General News: Microsoft Feels the Heat

AWS: A Very Fargate Indeed

This week’s highlights

General News: Great Expectations

AWS: Everything Generally Available

Related Podcasts

Links

Stay Connected