A conference that wasn’t about cloud

At the recently concluded Cloud Next 2024 conference, Google made a number of announcements highlighting its commitment to generative AI. The conference is typically associated with cloud infrastructure, platforms, and related services, but the theme of generative AI overshadowed everything else. Leaders representing various departments and organizations within Google Cloud highlighted how generative AI is being incorporated into their products and services.

The salient takeaway from Google Cloud Next 2024 is that generative AI now plays a pivotal role within Alphabet Inc, with Gemini playing a central role. This change highlights Google Cloud’s critical position in ushering in the tech giant’s next wave of innovation.

Gemini’s deep integration into Google’s ecosystem not only increases the strategic importance of Google Cloud, but also positions Google Cloud to power Alphabet’s future growth and technology advancements.

While there have been advances in core infrastructure services like Compute Engine, Kubernetes Engine, and Cloud Storage, the keynote was used to demonstrate how Google’s generative AI technology is superior to its competitors. . Compared to other hyperscalers’ keynotes, Generative AI, led by Gemini, stole the show.

There are five important points for companies:

Gemini brand consolidation and rationalization

After experimenting with multiple brands and messaging, Google has finally brought all of its generative AI services under one roof – Gemini. This new branding effort will help Google simplify its services across consumer, developer, and enterprise users. Gemini 1.5 Pro, the latest in multimodal AI, powers all products and services available through Google Cloud and Google Workspace.

Duet AI, a brand launched at the last Cloud Next event, was quickly replaced by Gemini. Duet AI for developers is now Gemini Code Assist, and Duet AI for Google Cloud is now Gemini Cloud Assist. Gemini for Google Workspace replaces Duet AI for Google Workspace. So it’s clear that Google wants to consolidate and streamline all of its generative AI efforts under Gemini.

Google calls PaLM 2 the predecessor of Gemini, which powers the Duet AI user interface. With the new branding, the brain, the language model, and the user-facing tools have the same name. By merging the brands, Google gave its AI a catchy and consumer-friendly name: “Gemini.” This is a strategic move to gain mindshare and recall of diverse target segments ranging from consumers to businesses.

AI agents are everywhere

Google is expanding its investment in generative AI through a customized vertically integrated AI agent powered by Gemini. Agents are built by Google partners and customers, leveraging intelligence provided by generative AI models and private data stored in cloud and on-premises environments. You will be able to mimic the functionality and automation capabilities provided by some of Google’s first-party agents, such as Gemini Cloud Assist for DevOps and Gemini in Security for DevSecOps.

Google has renamed its Vertex AI Search and Conversation service to Agent Builder. This enables customers to build intelligent automation that is tightly grounded in data. Google combines the benefits of Internet Search, Dialogflow, and Gemini into a single, unified no-code platform to design and deploy custom agents.

Agent is the answer to Microsoft Copilots, which provides similar functionality. Amazon has also integrated the ability to build custom agents into its generative AI platform, Bedrock.

Google will continue to enhance and deliver new agents under the Gemini brand, while also enabling partners and customers to build custom agents alongside Gemini-based automation tools.

Overall, the agent strategy is a good move that will create a new ecosystem and ultimately create a market for intelligent tools available to Google Cloud and Workspace customers.

Growing Gemini’s customer base

Google has gone out of its way to say that its AI is better than its competitors’ AI. This was evident from the number of customer stories presented in Thomas Kurian’s keynote address.

Generative AI is still in its infancy, but Google has an impressive lineup of companies ranging from Mercedes-Benz to Orange to IHG Hotels & Resorts to Best Buy to Bayer.

Bayer is developing a radiology platform that helps radiologists and other companies develop and deploy AI-first healthcare apps to improve efficiency and diagnostic turnaround times. Mercedes-Benz will collaborate with Google Cloud to use generative AI to improve customer-facing use cases in e-commerce, customer service, and marketing. Best Buy uses Gemini to create a new and convenient mechanism for customers to resolve product issues and reschedule deliveries.

The company examples and testimonials highlighted in the keynote certainly gave Google an edge.

Investing in custom silicon and AI hypercomputers

Google announced an AI hypercomputing strategy using high-speed computing hardware based on NVIDIA GPUs and in-house TPUs. The AI ​​hypercomputer is a performance-optimized infrastructure stack that leverages Google Cloud TPUs, Google Cloud GPUs, Google Cloud Storage, and the underlying Jupiter network to accelerate training of cutting-edge models at scale. .

Google is also investing in Axion processors, the first custom Arm-based CPUs designed for data centers. Axion offers industry-leading performance and energy efficiency and is expected to be available to customers later this year.

Axion processors combine Google’s silicon expertise with Arm’s highest-performing CPU cores to deliver up to 30% better performance than the fastest general-purpose Arm-based instances in the cloud and 50% better performance than comparable comparable instances. % better performance, and 60% more energy efficient. Current generation x86-based instances. Google has already started rolling out his Google services such as BigTable, Spanner, BigQuery, Blobstore, Pub/Sub, Google Earth Engine, and YouTube advertising platform to the current generation of Arm-based servers.

Google is also investing in Titanium, a proprietary data processing unit that offloads routine network, storage, and security operations to dedicated processors. The combination of Axion and Titanium provides customers with unparalleled scale, performance, and cost efficiency.

The Axion and Titanium chips are similar to what Amazon did with the AWS Nitro System and Graviton CPUs. Microsoft employs a similar strategy with its Azure Boost and Azure Cobalt processors.

Evolving your core infrastructure with storage and compute

Although the keynote didn’t mention advances and enhancements to core computing, networking, and storage infrastructure that are critical to cloud computing, Google has made significant improvements to its cloud infrastructure.

Google continues to focus on workload-optimized infrastructure through enhancements to Compute Engine virtual machines. First in the public cloud, Intel 5th Generation Xeon processors power the C4 and N4 families of general-purpose virtual machines. Powered by Titanium, these instances are designed to support general-purpose workloads by providing a balance of high performance, flexibility, and cost.

Introduced last year, Hyperdisk separates storage from compute to deliver higher throughput. Hyperdisk storage pools with advanced capacity will be announced at Next. In a typical scenario, customers manage block storage capacity and utilization in the cloud on a disk-by-disk basis, which is complex, labor-intensive, error-prone, and often results in underutilized resources. there is. Hyperdisk storage pools allow customers to purchase and manage block storage capacity in a pool that is shared among workloads. These pools are used to thinly provision individual volumes, using capacity only when data is actually written to disk, and leveraging data reduction techniques such as compression and deduplication.

Google Kubernetes Engine, a managed Kubernetes service, is optimized to host the underlying model and LLM. Customers can add secondary disks to nodes with the model preloaded. This significantly speeds up the service delivery process.

summary

Cloud Next 24 reveals Google aggressively pursuing AI dominance with significant investments across a broad technology stack and a clearly focused, unified brand strategy based on Gemini . Google’s commitment to generative AI was undeniably positioned as the cornerstone of its competitive advantage in an evolving technology environment.