About this episode
Welcome to episode 255 of the Cloud Pod podcast – where the forecast is always cloudy! This week your hosts, Justin, Jonathan, Matthew and Ryan are here to tackle the aftermath of Google Next. Whether you were there or not, sit back, relax, and let the guys dissect each day’s keynote and the major announcements. Titles we almost went with this week: How About Some AI? ⛅“The New Way to Cloud” is a Terrible TagLine (and is what happens when you let AI do your copy) Welcome Google Cloud Next Where There is No Cloud, Just AI Ok Google, did your phone go off? For 100 dollars, guess how many AI stories Google Has This Week From Search to Skynet: Google Cloud Next’s Descent into AI Madness ‘Next’ Up from Google – AI! Have Some Conference with Your AI A big thanks to this week’s sponsor: We’ve got a new sponsor! Sonrai Security Check out Sonrai Securities’ new Cloud Permission Firewall. Just for our listeners, enjoy a 14 day trial at sonrai.co/cloudpod GCP – Google Next 2024 We’re jumping right into GCP this week, so we can talk about all things Google Next. 01:44 FIrst impressions: Vegas > Moscone, so take that Vegas. Both Ryan and Justin agree that Vegas is much better than the Mosconoe center in San Francisco for Google Next The Sessions were well organized, but Ryan is a little tired from walking back and forth between them. Exercise is tiring! \ Vegas infrastructure was well utilized, something Amazon didn’t do as well. Folks staying at area hotels that *weren’t* Mandalay Bay had some issues with trying to get onto / off property at the beginning and end of the day. Free coffee is still available. *If you can find it. Expo hall felt cramped 08:22 Thoughts on the Keynote Address Note: Not enough space in the arena for keynotes; the arena holds approx. 12k; numbers released by Google say there were 30k in attendance. Thomas Kurian kicked off the keynote, introduced their new tagline “The New Way to Cloud” Sundar: Months can feel like decades in the cloud… WORD. 36B revenue run rate Kurian did a rapid fire announcement of all the things coming – which required Justin to rewatch just to get them all. A3 Mega Nvidia H100 GPUs Nvidia GB200 NVL72 (in early 2025 TPU v5p GA Hyperdisk ML for Inference Cloud Storage Fuse Caching GA Parallel Store Caching AI Hypercomputer Dynamic Workload Scheduler Nvidia GPU Support for GDC Google Distributed Cloud GKE Enterprise for GDC AI Models on GDC Vector Search on GDC Vertex AI Solutions with GDC Secret and Top Secret Accreditations Via the Price is Right Door TK: Google Axion Processor first custom ARM processor CPU Gemini 1.5 Pro to public preview CodeGemma lightweight coding focused LLM Supervised tuning for Gemini Models Grounding with Google Search Automatic Side by Side (AutoSxS) Rapid Evaluation Vector Search GA for customer agent Gemini for Google Workspaces (Rebrand on Duet) AI Meetings and messaging add on AI Security Add on Gemini in Chat (So now you can avoid participating in the chat, and just get a recap.) Imagen 2.0 GA in Vertex AI Text to Live Image Digital Watermarking for images created by Imagen Now editing modes in imagen 2.0 Gemini in Bigquery Bigquery Data Canvas (Basically gives you a journal-type interface) Vector Indexing in Bigquery and AlloyDB Direct access to Vertex AI from BigQuery Gemini in Looker – you can ask it questions about the data your looking at Duet AI is now Gemini Code Assist Gemini 1.5 Pro is being added to Code Assist (gives you access to 1 million tokens – meaning it can process 6 hours of audio / 1 hour of video ) Gemini Cloud Assist to help across your application lifecycle Gemini in Threat Intelligence Gemini in Security Operations Gemini in Security Command Center – which Ryan got to play with a little. 19:45 Ryan – “I got access to an early sort of hands – on lab and it was a really powerful way to look at your data, because you can pipe so much data into the security tools, and it becomes a little bit daunting to go and figure out what’s going on and to proactively monitor and feel like you’re actually catching things, to make certain patterns. And so the ability to sort of use prompt to query data, and have it generate certain query codes within to generate reports or building lists was very powerful, very cool. And it made it a lot of fun. It was probably the most fun hands – on lab I’ve ever done; just by seeing, you know, it’s all dummy data, but it was a fun way to track things through for forensic analysis and to understand patterns and life cycle. Super impressed.” If you’re not keeping track – that was 41 items announced in a 90 minute keynote. And even with all that – Justin and Ryan struck out on their predictions. Jonathan and Matthew are tied at 1 correct prediction each. Justin Gemini 2.0 will be announced and available at Google Next LLM/Prompt Security from Mandian/Google solution Something around the brand confusion of GKE Enterprise/Anthos Jonathan IAM Conditions to support calling a web service or cloud function to do more dynamic permissions Anything for IAM on the mainstage A new Security Forensics capability (threat hunting/Siem/ish) Matt Healthcare company will be a guest on the mainstage with all the back end processing etc. Integrations with Gemini and BigQuery GCP will make fun of Azure for backing Redis in a subtle way Ryan GCP will highlight how their use AI to tackle the climate crisis GCP will announce a managed NFS/CIFS solution. (EFS competitor) GCP will announce an AI enhancement or robot to their google assistant homepod Google Next Tie Breaker: How many times will they say AI/LLM on stage? 111 Ryan – 67 Matt – 142 Jonathan – 52 Justin – 78 Number of main stage announcements? 41 Matt – 25 Jonathan – 9 Justin – 1 Ryan – 2 Congratulations to Matthew, who wins Google Next ‘24 Predictions. 24:32 Ryan – “So the one thing I will say is they pushed a lot of things that weren’t AI product related into the sessions themselves.S o one thing I learned this year that I’ll take forward to other conferences is the what’s new in blah sessions. So, you know, I wanted to go, I tried to get in the one for VPC networking and I wasn’t able to get in, but I did manage to get into like IAM and a couple others. So there were a lot of announcements for enhancements done directly in those.” 25:20 Justin – “So other day one announcements that did not make main stage are still also pretty nice. They have new computing and networking capabilities, including a new C4 and N4 general purpose VMs powered by the fifth generation Intel Xeon processor, as well as those enhancements to Google Cloud we talked about earlier. There were a bunch of database enhancements that didn’t get to the main stage, including AlloyDB AI, Firestore enhancements for flexible natural language support, as well as vector capabilities for Firestore as well.” 25:20 Jonathan – “Yeah, I hope AI just becomes one of those features that everything has and they don’t devote a whole keynote to it again because I think it was a little much. It was really hard, you know, sitting listening at home. It was honestly too fast paced, too many announcements for keynote in, you know, in 90 minutes and everything was just AI, AI, AI the whole time. And I kind of tuned out honestly, because yes, I mean, once you’ve seen one AI integration with a tool, you can kind of guess what all the others are going to look like. So, you know, either have a separate AI conference to focus on the tech, but perhaps not the use cases for it all the time. But I’m kind of hoping they’ll go back to actually talking about the rest of the ecosystem and the work they’re doing there as well.” 30:38 Day 2 – Developer Keynote New appHub New BigQuery continuous query New natural language support in AlloyDB Gemini Code Assist in Apigee API management Detecting shadow API usage in Apigee Cloud Run application canvas Gen AI Quick Start Solutions for GKE Support for Gemma on GKE Vertex AI MLOps capabilities Shadow API detection Confidential Accelerators for AI workloads GKE container and model preloading 32:44 Jonathan – “ the TPU support for GKE should have been a bit more… I mean, I think they mentioned it in the keynote, didn’t they? But it didn’t seem to get much attention, but that’s really useful. I mean, that’s sort of commoditizing training, which it hasn’t been before.” 34:25 Finops Announcements Finops billing data to become more timely – they reduced the time required for data updates by 30%, giving you a faster understanding of what you’re spending, which is always a good thing. Cloud Billing announced support for cloud storage costs at the bucket level and storage tags. New cost anomaly detection ability – no setup required. New management capabilities for dealing with committed use discounts and viewing those in the FinOps hub Need some links? We have some for ‘ya! Welcome to Google Cloud Next ‘24 Day 1 at Next ’24 recap: AI agents for everyone Day 2 at Next ’24 recap: building AI agents Cloud FinOps news from Next ‘24 April 11, 2024 Run AI anywhere with Google Distributed Cloud innovations Introducing Google Axion Processors, our new Arm-based CPUs Make Google part of your security team anywhere you operate, with defenses supercharged by AI What’s next for data analytics at Google Cloud Next ’24 What’s new and what’s next for Google Cloud databases What’s new with Google Cloud Networking at Next ’24 What’s new in Google Cloud’s workload-optimized infrastructure Powering Google Cloud with Gemini What’s new with Google Cloud’s AI Hypercomputer architecture Privacy-preserving data sharing now generally available with BigQuery data clean rooms Ushering in a new era for app developers Google Public Sector achieves Top Secret and Secret cloud authorization Announcing Vertex AI Agent Builder: Helping developers easily build and deploy gen AI experiences Introducing Chrome Enterprise Premium: The future of endpoint security Google Cloud announces updates to Gemini, Imagen, Gemma and MLOps on Vertex AI The container platform for the next decade of AI and beyond Grounding generative AI in enterprise truth Analyze images and videos in BigQuery using Gemini 1.0 Pro Vision How Gemini in BigQuery accelerates data and analytics workflows with AI Accelerate AI Inference with Google Cloud TPUs and GPUs Gemini in Databases — supercharge database development and management BigQuery is now your single, unified AI-ready data platform Introducing Gemini in Looker to bring intelligent AI-powered BI to everyone Get to know BigQuery data canvas: an AI-centric experience to reimagine data analytics Celebrating 20 years of Bigtable with exciting announcements at Next New Google Cloud Consulting programs designed to accelerate your cloud journey Announcing Cloud Service Mesh – the evolution of service mesh for Google Cloud Using Gemini Code Assist to build APIs, integrations, and automation flows Introducing Shadow API detection for your Google Cloud environments Natural language support in AlloyDB for building gen AI apps with real-time data Introducing ScaNN vector indexing in AlloyDB, bringing 12 years of Google research to speed up vector search Expanded Confidential Computing portfolio and introducing Confidential Accelerators for AI workloads Gemma on Google Kubernetes Engine deep dive: New innovations to serve open generative AI models Powering generative AI with cloud storage innovations at Next ’24 App Hub – Manage your application, forget the toil Performance deep dive of Gemma on Google Cloud Introducing ML Productivity Goodput: a metric to measure AI system efficiency Your scannable list of our top migration announcements from Next ‘24 Eating our own dogfood: Building an AI-driven business at Google Cloud Consulting Turbocharge applications with Memorystore’s persistence and flexible node types Private, secure, and seamless connectivity to Cloud SQL using Private Service Connect What’s new with Firestore at Next ‘24 Announcing the general availability of Next Gen Firewall Enterprise Migrate your SQL Server workloads to Cloud SQL with Database Migration Service, now in preview Accelerating database modernization with Gemini in Database Migration Service Introducing Isolator: Enabling secure multi-party collaboration with healthcare data Announcing Delta Lake support for BigQuery Build powerful gen AI applications with Firestore vector similarity search Closing And that is the week in the cloud! Go check out our sponsor, Sonrai and get your 14 day free trial. Also visit our website, the home of the Cloud Pod where you can join our newsletter, slack team, send feedback or ask questions at theCloud Pod.net or tweet at us with hashtag #theCloud Pod