Google Cloud Next’24
Google Cloud kicked off its annual Next event in Las Vegas, on April 9th, 2024. It was a three-day event packed with new product releases and advancements, hands-on demos, and insightful customer stories. As usual, it was devided into:
The Opening Keynote – featuring new product releases across Google Cloud’s offerings and how customers are building solutions on Google Cloud.
The Developer Keynote – featuring demos on how developers can leverage these new technologies to build better on Google Cloud.
So much went down during this event and here is a summary of the announcements highlighting how Google Cloud is delivering on their vision of a new way to cloud.
Gemini for Google Cloud
From the numerous announcements made at Cloud Next ‘24, Google’s most capable AI model yet, Gemini, took center stage. Following Gemini 1.5 Pro’s release back in February, Google Cloud announced three new services powered by its latest version of Gemini.
Gemini Cloud Assist – Helps you to manage your app’s entire lifecycle on the cloud. It understands your cloud environments and offers recommendations to help you design new apps, troubleshoot issues, and optimize performance.
Gemini Code Assist – An AI code assistant that completes code as you write and writes entire code blocks on demand. It is available in popular IDEs including VS Code, JetBrains IDEs, and more.
Gemini in Security Operations – Empowers security investigations in Google Security Operations. It helps run searches targeted toward specific events, answer security questions, and suggest follow-on steps using natural language prompts.
Apart from these, Google also announced that Gemini has been integrated into other services on GCP, including BigQuery, Google Workspace, etc.
A New App, Better Security, and Increased Efficiency in Google Workspace
Google Cloud announced a wave of innovations coming to Google Workspace at Cloud Next ‘24, beginning with the AI Meetings and Messaging Add-on. This add-on can take notes for you and translate your meetings in real-time. Secondly, to introduce AI-powered data protection in Google Workspace, Google Cloud announced the general availability of the AI Security Add-on.
Gemini has made waves in Google Workspace resulting in Gemini in Google Chat. This service can summarize your chats to catch you up on long conversation threads.
Google Vids
In one of the biggest announcements made during the event, Google released the newest Workspace app, Google Vids. It is an AI-powered video creation and editing app for work. Using natural language prompts, teams will be able to create any type of video for work, including sales training videos and product demos.
New Infrastructure to Support Google’s Work with AI
Google Cloud announced the general availability of Cloud TPU v5p and the AI Hypercomputer. Cloud TPU v5p is currently Google’s most powerful and flexible AI accelerator, capable of supporting the training of powerful AI models.
Cloud TPU v5p powers the AI Hypercomputer, an accelerated supercomputer architecture for AI workloads. This architecture improves the efficiency of training, tuning, and serving AI models by employing performance-optimized hardware, open software, and leading ML frameworks.
In addition, Google announced the upcoming general availability of A3 Mega VMs, powered by NVIDIA’s H100 tensor core GPUs. This will be followed by NVIDIA’s Grace Blackwell generation of GPUs, coming to Google Cloud in early 2025.
What’s New in Vertex AI?
Vertex AI took a bit of a step back in Google Cloud Next ‘24, compared to Next ‘23. Nevertheless, here is what Google announced about its unified AI development platform.
Gemini 1.5 Pro is available for public preview in the Vertex AI Model Garden. Users can try it out and provide feedback on how well it works and how it can be improved.
Vertex AI’s grounding capabilities have been expanded. Google has made it easy for users to directly ground the responses of their models with Google Search and enterprise data.
The Model Garden keeps expanding with new models, the latest addition being Code Gemma. A new version of Imagen, Imagen 2.0 has also been added.
The Agent Builder
Google has expanded and rebranded Vertex AI Search and Conversation into the Agent Builder. This service allows you to build AI agents – smart applications that perceive their environment and take action to achieve specific goals. For instance, customer service chatbots that respond to customer queries and provide support.
Core Infrastructure and Networking Announcements
Google announced its new processor, Google Axion, in perhaps the biggest announcement made during the event. This is an Arm-based processor built for data center operations like web and app servers, microservices, etc.
For networking, Google revealed that it is combining the capabilities of its Traffic Director with Anthos Service Mesh to form Cloud Service Mesh. In this service, users get to access the Traffic Director’s control plane to efficiently and scalably route traffic in their service meshes.
Gemini in Databases
BigQuery is among the GCP services receiving a boost thanks to its integrations with Gemini. Gemini in BigQuery works to simplify the journey from raw data to insights through:
AI-augmented data preparation to increase the speed of cleansing data.
Allowing users to build low-code data pipelines or replace legacy pipelines in BigQuery.
Apart from Gemini in BigQuery, Google also announced the continuous query feature, which is still under development. You can use it to run specific types of queries continuously by updating them in real-time as new data arrives in the source tables.
Currently, in preview, Google Cloud is also working on developing a link to allow users to access AI models in Vertex AI directly from BigQuery. This will allow you to leverage AI models in the Model Garden to analyze your data in BigQuery.
Gemini in Data Analytics
In this age of big data. a data mesh offers a compelling approach to data management. It empowers domain teams to create high-quality data for consumption, fosters collaboration, and ultimately helps enterprises unlock the true value of their data.
Google’s Data Cloud empowers businesses to seamlessly build a data mesh architecture for their data. It provides innovative tools that facilitate decentralized data ownership, treating data as a product, and creating governance policies to be applied across the data mesh. It also allows enterprises to bring together different services to create a self-serve data platform to make this data easily accessible across the organization.
Security
Since 2022, Google has been using Mandiant’s threat intelligence. Mandiant in Google Security Operations equips organizations with early threat insights and response expertise for high-profile incidents. With Gemini in Threat Intelligence, users can now access Mandiant’s threat intelligence with simple conversational text prompts.
Gemini in Security Command Center works similarly to allow you to search for security events. It also summarizes vulnerabilities and attack paths for you. Apart from Gemini in Security Command Center, Google also announced that Security Command Center Enterprise is now generally available. This a SecOps solution designed to help users stay secure across multiple clouds, mitigating multi-cloud risk.
Lastly, Google has added new features to improve its browser’s security with Chrome Enterprise Premium. Chrome Enterprise premium adds data protection with the data loss prevention API and Zero Trust security controls. It also allows organizations to set their own policy controls on Google Chrome and get security insights and reporting on usage.
Summary and Key Takeaways
The announcements made at Cloud Next ‘24 highlight Google’s heavy focus on integrating Gemini, its most powerful AI model yet, across its offerings. Gemini has been integrated into almost all of Google Cloud’s services, from its collaboration apps, data storage, and data analytics to security solutions. On top of this, it continues to re-invent and innovate its infrastructure to enhance its work with AI and follow through with its vision of becoming an AI-first company.