Kategorie: Google Cloud Platform

London Summit: UK businesses turn to Google Cloud AI

The AI era is here, and the UK is at the forefront. 
Over the past year, search interest for “AI” has surged by 50% in the country, while inquiries about “how to use AI” have jumped 70%. By 2030, AI could generate more than £400 billion ($524 billion) in value for the UK economy and save more than 700,000 administrative hours annually across sectors like healthcare, education, and finance. Central to this transformation is Gemini, Google Cloud’s next-generation AI foundation model, which is empowering UK businesses across all sectors to unlock new possibilities in a rapidly evolving landscape.
This week, the energy is palpable, as we welcome customers and partners to the Google Cloud Summit London at Tobacco Dock to experience how Gemini is helping shape the future of business and unlocking business value. The summit also showcases innovations from across Google Cloud’s extensive portfolio of products and services, as well as from our ecosystem of customers and partners — illustrating how AI is transforming different industries and equipping UK businesses to stay ahead in a competitive landscape.
And what a way to kick off the summit than with yesterday’s announcement of Google’s deepening partnership with Vodafone. This expands beyond our longstanding work together in the cloud, with a 10-year, billion-plus-dollar deal that includes cybersecurity, devices, and cloud services across Europe and Africa. As Google’s CEO, Sundar Pichai, noted: “I’m excited to see how Vodafone’s consumers, small businesses, and governments will use generative AI and Google Cloud to transform the way they work and access information.”
I’m also excited to further support our customers’ growth with Google Cloud’s $1 billion investment in a new UK data center in Waltham Cross, Hertfordshire. Unveiled in January, this major project reinforces our commitment to AI innovation and delivering reliable digital services across Europe and the world. We’re also fueling this digital growth with key updates and announcements made today from both Google Cloud and our customers. 
Empowering customers with data and AI solutions
The momentum behind AI is undeniable — nearly two-thirds of UK organizations have earmarked at least half of their AI budget for generative AI. These tools have the potential to generate £4.8 billion in productivity savings, according to research from Public First.
Today, we are proud to highlight leading UK companies who are already harnessing Gemini and Google Cloud AI capabilities to drive real-world impact: 

BUPA UK is using Google Cloud to build a cloud-first digital customer platform providing direct access to care when needed, as well as the tools to enable proactive health outcomes for customers. This new digital platform will allow BUPA to develop innovative ways to improve members’ wellbeing, reducing the cost of care all while delivering a highly personalized healthcare experience.

Dunelm has partnered with Google Cloud to enhance its online shopping experience with a new gen AI-driven product discovery solution. This has shown significant improvements in a number of key areas, including reduced search friction, helping customers find the products they are looking for.

Incubeta is using Gemini for Google Workspace, primarily with their development teams, to streamline architecture documentation and code refinements, and to reduce time spent on these tasks. They also use it for client teams to assist with mundane tasks and for creative conception.

Vodafone — in addition to its 10-year expanded partnership with Google — also recently completed the migration of its SAP ERP resource planning database to Google Cloud. One of the largest shifts of its kind, it was completed with zero business interruptions. At the summit, the team will be sharing more about this work, which boosted performance while shrinking costs and Vodafone’s carbon footprint.

Data residency for machine learning available in the UK
We understand the importance of data sovereignty for certain industries. That’s why Google Cloud is expanding our data residency commitment, enabling UK organizations to run machine learning processing for Gemini 1.5 Flash within the UK. 
This effort helps address key data sovereignty and compliance concerns that enable UK organizations, including the public sector, to leverage gen AI while maintaining strict control over their data. 
Fueling the next generation of UK and EMEA startups
We are also excited to share that since 2023, more than 60% of UK-based gen-AI startups are Google Cloud customers. Yesterday, at our global Startup Summit, we announced several important updates that are relevant for UK AI startups. These include our Startup School for AI skill-building and the ISV Startup Springboard offering extensive go-to-market and co-selling opportunities with Google Cloud and our partners. We’ve also launched a partnership with the startup accelerator 500 Global to provide streamlined access to Google Cloud’s AI platform
As part of today’s summit, we’re launching the Google Cloud Startup Hub in London — a dedicated community space designed to provide in-person education and engagement for startups and developers. Open five days a week with business hours from 9 a.m. to 6 p.m. or until 9 p.m. to accommodate evening events, the hub offers hands-on learning for Google Cloud and partner solutions and access to experts. 
At the hub, we will host exclusive sessions with industry leaders and facilitate invaluable networking opportunities with investors and peers. The Google Cloud Startup Hub also serves as a vital community resource where key partners can also host events. 
Some examples of how UK startups are driving breakthroughs with Google Cloud are: 

BioCorteX, in collaboration with Google Cloud, announces a breakthrough in ADC research for cancer treatment using its “Unified Biology” approach. The research has uncovered a crucial link between the tumor microenvironment and ADC efficacy, potentially transforming personalized cancer therapy. 

Motorway uses Google Cloud AI to streamline the process of buying and selling used cars online. With Google Cloud’s AI, Motorway has been able to build and deploy AI models faster, automate document processing, and provide industry-leading vehicle valuations.

OnBuy, a fast-growing ecommerce marketplace, partners with Google Cloud for international expansion and enhanced customer experience. Google Cloud’s infrastructure, data storage, and AI tools enable OnBuy to scale, improve efficiency, and innovate.

We’re also announcing the opening of the new AI Playground, located within the same space as the Google Cloud Startup Hub. An experiential AI demo space, the AI Playground is aimed at inspiring and empowering developers and organizations to build incredible new applications and solve business challenges with AI and ML.
Google Cloud’s commitment to making data and AI more accessible and powerful for all users
To help enhance our overall data platform, we’re also sharing a number of important product updates and announcements. These will deliver better insights and efficiencies for our customers and boost the performance of their AI offerings and projects. Data is the foundation of any AI work, so without having a handle on the former, it’s hard to succeed in the latter.
BigQuery, our unified enterprise gen AI-ready data platform, has now been integrated with our Gemini models. Gemini in BigQuery makes data prep easier, offering an intuitive natural-language interface that can help data teams generate insights from metadata and other large datasets, and it’s especially useful for new batches of data. 
We also have new synthetic data capabilities, with BigQuery Dataframes, which make it easier to run and train models when inputs are more limited. Gemini in Looker has added conversational analytics, which lets you use a search experience similar to those found across Google; this allows for data exploration using natural language that surfaces actionable insights more easily.
For enhanced security, BigQuery now supports cross-region disaster recovery with multi-factor authentication, ensuring business continuity and data protection. 
New managed BigQuery workflows assist data engineers in building data pipelines. For data ingestion, customers can use new managed services for Flink and Kafka to better configure, tune, scale, monitor, and upgrade real-time workloads. A unified data catalog in BigQuery helps organize and manage all your data and metadata, making it easier to discover and use, while BigQuery catalog semantic search, available in preview, expands the ability to find data using natural language. These search features make BigQuery more intuitive and accessible to everyone.
Google is also launching a new enterprise tier of Code Assist, our AI-powered coding partner, which can help teams with work throughout the software development lifecycle. This enterprise tier will offer enhanced security, improved context for better accuracy and reliavbility, and wider integration with Google Cloud services.
With these investments and the customer momentum showcased today, Google Cloud is reinforcing the UK’s role in shaping the future of AI and strengthening the future we can build together on the cloud. 
Call to action – Browse upcoming events & sign up!
Quelle: Google Cloud Platform

Your ultimate guide to the latest in generative AI on Vertex AI

The world of generative AI is evolving at a pace that’s nothing short of mind-blowing. It feels like just yesterday we were marveling at AI-generated images, and now we’re having full-fledged conversations with AI chatbots that can write code, craft poetry, and even serve our customers (check out our list of 101 real-world gen AI use cases from the world’s leading organizations).
The pace of innovation can be hard to keep up with — in 2023 we introduced over 500 new features in Vertex AI and we’re not slowing down this year. We put this blog together to help you keep track of the biggest announcements and make sense of what they mean for your business. We’ll keep it updated as new announcements come out, so bookmark this link and check out our new video series below also covering new announcements. 
Catch up on the latest announcements
We recently announced several updates to make Gemini, Meta and Mistral models more accessible, followed shortly by announcing the Jamba 1.5 Model Family from AI21 Labs.

Lama 3.1 & Mistral AI on Vertex AI

Lower pricing for Gemini 1.5 Flash

What it is: We’ve updated Gemini 1.5 Flash to reduce the input costs by up to ~85% and output costs by up to ~80%, starting August 12th, 2024. 

Why it matters: This is a big price drop on Gemini Flash, a world-class model with a 1 million context window and multi-modal inputs. Plus, coupled with capabilities like context caching you can significantly reduce the cost and latency of your long context queries. Using Batch API instead of standard requests can further optimize costs for latency insensitive tasks. 

Get started: View pricing to learn more and try out Gemini 1.5 Flash today.

More Languages for Gemini

What it is: We’re enabling Gemini 1.5 Flash and Gemini 1.5 Pro to understand and respond in 100+ languages. 

Why it matters: We’re making it easier for our global community to prompt and receive responses in their native languages.

Get started: View documentation to learn more.

Meta’s Llama 3.1 

What it is: Llama 3.1 models are now available on Vertex AI as a pay as you go API, this includes 405B, 70B and 8B (coming in early September). 

Why it matters: 405B is the largest openly available foundation model to date. 8B and 70B are also new versions that excel at understanding language nuances, grasping context, and performing complex tasks such as translation and dialogue generation. You can access the new models in just a few clicks using Model-as-a-Service, without any setup or infrastructure hassles. 

Get started: To access Llama 3.1, visit Model Garden

Mistral AI’s latest models

What it is: We added Mistral Large 2, Nemo and Codestral (Google Cloud is the first hyperscaler to introduce Codestral). 

Why it matters: Mistral Large 2 is their flagship model which offers their best performance and versatility to date and Mistral Nemo is a 12B model that delivers exceptional performance at a fraction of the cost. Codestral is Mistral AI’s first open-weight generative AI model explicitly designed for code generation tasks. You can access the new models in just a few clicks using Model-as-a-Service, without any setup or infrastructure hassles. 

Get started: To access the Mistral AI models, visit Model Garden (Codestral, Large 2, Nemo) or check out the documentation

Jamba 1.5 Model Family from AI21 Labs

What it is: Jamba 1.5 Model Family  — AI21 Labs’ new family of open models — is in public preview on Vertex AI Model Garden, including:  

Jamba 1.5 Mini: AI21’s most efficient and lightweight model, engineered for speed and efficiency in tasks including customer support, document summarization, and text generation.

Jamba 1.5 Large: AI21’s most advanced and largest model that can handle advanced reasoning tasks — such as financial analysis — with exceptional speed and efficiency. 

Why it matters: AI21’s new models join over 150 models already available on Vertex AI Model Garden, further expanding your choice and flexibility to choose the best models for your needs and budget, and to keep pace with the continued rapid pace of innovation. 

Get started: Select the Jamba 1.5 Mini or Jamba 1.5 Large model tile in Vertex AI Model Garden. 

Previous Announcements
Best models from Google and the industry
We’re committed to providing the best model for enterprises to use – Vertex AI Model Garden provides access to 150+ models from Google, Partners and the open community so customers can select the model for the right price, performance, and latency considerations.
No matter what foundation model you use, it comes with enterprise ready tooling and integration to our end to end platform. 
Gemini 1.5 Flash is GA

What it is: Gemini 1.5 Flash combines low latency, highly competitive pricing, and our 1 million-token context window.

Why it matters: Gemini 1.5 Flash is an excellent option for a wide variety of use cases at scale, from retail chat agents, to document processing, to research agents that can synthesize entire repositories.

Get started: Click here to get started now with Gemini 1.5 Flash on Vertex AI. 

Gemini 1.5 Pro, GA with 2-million -token context capabilities 

What it is: Now available with an industry-leading context window of up to 2 million tokens, Gemini 1.5 Pro is equipped to unlock unique multimodal use cases that no other model can handle.

Why it matters: Processing just six minutes of video requires over 100,000 tokens and large code bases can exceed 1 million tokens — so whether the use case involves finding bugs across countless lines of code, locating the right information across libraries of research, or analyzing hours of audio or video, Gemini 1.5 Pro’s expanded context window is helping organizations break new ground. 

Get started: Click here to get started now 

Imagen 3 is GA

What it is: Google’s latest image generation model, delivering outstanding image quality, multi-language support, built-in safety features like Google DeepMind’s SynthID digital watermarking, and support for multiple aspect ratios.

Why it matters: There are several improvements over Imagen 2 — including over 40% faster generation for rapid prototyping and iteration; better prompt understanding and instruction-following; photo-realistic generations, including of groups of people; and greater control over text rendering within an image. 

Get started: Apply for access to Imagen 3 on Vertex AI. 

Gemma 2

What it is: The next generation in Google’s family of open models built to give developers and researchers the ability to share and commercialize their innovations, using the same technologies used to create Gemini. 

Why it matters: Available in both 9-billion (9B) and 27-billion (27B) parameter sizes, Gemma 2 is much more powerful and efficient than the first generation, with significant safety advancements built in. 

Get started: Access Gemma 2 on Vertex AI here.

Anthropic’s Claude 3.5 Sonnet

What it is: We recently added Anthropic’s newly released model, Claude 3.5 Sonnet, to Vertex AI. This expands the set of Anthropic models we offer, including Claude 3 Opus, Claude 3 Sonnet, Claude 3 Haiku. You can access the new models in just a few clicks using Model-as-a-Service, without any setup or infrastructure hassles. 

Why it matters: We’re committed to empowering customer choice and innovation through our curated collection of first-party, open, and third-party models available on Vertex AI. 

Get started: Begin experimenting with or deploying in production Claude 3.5 Sonnet on Vertex AI.

End-to-end model building platform with choice at every level
Vertex AI Model Builder enables you to build or customize your own models, with all the capabilities you need to move from prototype to production.
Lower cost with context caching for both Gemini 1.5 Pro and Flash

What it is: Context caching is a technique that involves storing previous parts of a conversation or interaction (the “context”) in memory so that the model can refer back to it when generating new responses

Why it matters: As context length increases, it can be expensive and slow to get responses for long-context applications, making it difficult to deploy to production. Vertex AI context caching helps customers significantly reduce input costs, by 75 percent, leveraging cached data of frequently-used context. Today, Google is the only provider to offer a context caching API. 

Get started: Learn more in documentation.

Controlled generation

What it is: Controlled generation lets customers define Gemini model outputs according to specific formats or schemas. 

Why it matters: Most models cannot guarantee the format and syntax of their outputs, even with specified instructions. Vertex AI controlled generation lets customers choose the desired output format via pre-built options like YAML and XML, or by defining custom formats. 

Get started: Visit documentation to learn more.  

Batch API

What it is: Finally, batch API is a super-efficient way to send large numbers of non-latency sensitive text prompt requests, supporting use cases such as classification and sentiment analysis, data extraction, and description generation. 

Why it matters: It helps speed up developer workflows and reduces costs by enabling multiple prompts to be sent to models in a single request.

Get started: View documentation to get started.

New model monitoring capabilities

What it is: The new Vertex AI Model Monitoring includes

Support for models hosted outside of Vertex AI (e.g. GKE, Cloud Run, even multi-cloud & hybrid-cloud)

Unified monitoring job management for both online and batch prediction

Simplified configuration and metrics visualization attached to the model, not the endpoint

Why it matters: Vertex AI’s new model monitoring features provide a more flexible, extensible, and consistent monitoring solution for models deployed on any serving infrastructure (even outside of Vertex AI, e.g. Google Kubernetes Engine, Cloud Run, Google Compute Engine and more).

Get started: Learn more in this blog.

Ray on Vertex AI is GA

What it is: Ray provides a comprehensive and easy-to-use Python distributed framework. With Ray, you configure a scalable cluster of computational resources and utilize a collection of domain-specific libraries to efficiently distribute common AI/ML tasks like training, serving, and tuning. 

Why it matters: This integration empowers AI developers to effortlessly scale their AI workloads on Vertex AI’s versatile infrastructure, which unlocks the full potential of machine learning, data processing, and distributed computing.

Get started: Ready the blog to learn more.

Prompt Management

What it is: Vertex AI Prompt Management, now in preview, provides a library of prompts for use among teams, including versioning, the option to restore old prompts, and AI-generated suggestions to improve prompt performance. 

Why it matters: This feature makes it easier for organizations to get the best performance from gen AI models at scale, and to iterate more quickly from experimentation to production. Customers can compare prompt iterations side by side to assess how small changes impact outputs, and the service offers features like notes and tagging to boost collaboration. 

Get started: Visit documentation to learn more.

Evaluation Services 

What it is: We now support Rapid Evaluation in preview to help users evaluate model performance when iterating on the best prompt design. Users can access metrics for various dimensions (e.g., similarity, instruction following, fluency) and bundles for specific tasks (e.g., text generation quality). We also launched RAG and Grounded Generation evaluation metrics for summarization and question answering (eg: groundedness, answer_quality, coherence). For a side by side comparative evaluation, AutoSxS is now generally available, and helps teams compare the performance of two models, including explanations for why one model outperforms another and certainty scores that help users understand the accuracy of an evaluation.

Why it matters: Evaluation tools in Vertex AI help customers compare models for a specific set of tasks in order to get the best performance. 

Get started: Learn more in documentation.

Develop and deploy agents faster, grounded in your enterprise truth
Vertex AI Agent Builder allows you to easily and quickly build and customize AI Agents – for any skill level. A core component of the Vertex AI Agent Builder is Vertex AI Search, enabling you to ground the models in your data or the web. 
Grounding at Vertex
You have many options for Grounding and RAG at Vertex. These capabilities address some of the most significant hurdles limiting the adoption of generative AI in the enterprise: the fact that models do not know information outside their training data, and the tendency of foundation models to “hallucinate,” or generate convincing yet factually inaccurate information. Retrieval Augmented Generation (RAG), a technique developed to mitigate these challenges, first “retrieves” facts about a question, then provides those facts to the model before it “generates” an answer – this is what we mean by grounding. Getting relevant facts quickly to augment a model’s knowledge is ultimately a search problem.  
Read more at this blog post.
Grounding with Google Search is GA

What it is: When customers select Grounding with Google Search for their Gemini model, Gemini will use Google Search, and generate an output that is grounded with the relevant internet search results. Grounding with Google Search also offers dynamic retrieval, a new capability to help customers balance quality with cost efficiency by intelligently selecting when to use Google Search results and when to use the model’s training data. 

Why it matters: Grounding with Google Search is simple to use and makes the world’s knowledge available to Gemini.  Dynamic retrieval will save you money and will save your users time, only grounding when needed.

Get started: Read documentation to learn more about how to get started.

Grounding with third-party datasets

What it is: Vertex AI will offer a new service that lets customers ground their models and AI agents with specialized third-party data. We are working with providers like Moody’s, MSCI, Thomson Reuters and Zoominfo to enable access to their datasets.

Why it matters: These capabilities will help customers build AI agents and applications that offer more accurate and helpful responses. 

Get started: Coming soon, contact sales to learn more. 

Grounding with high-fidelity mode 

What it is: High-fidelity mode is powered by a version of Gemini 1.5 Flash that’s been fine-tuned to only use customer-provided content to generate answers and ensures high levels of factuality in response. 

Why it matters: In data-intensive industries like financial services, healthcare, and insurance, generative AI use cases often require the generated response to be sourced from only the provided context, not the model’s world knowledge. Grounding with high-fidelity, announced in experimental preview, is purpose-built to support such grounding use cases, including summarization across multiple documents, data extraction against a set corpus of financial data, or processing across a predefined set of documents.

Get started: Contact sales to learn more. 

Expanding Vector Search to support hybrid search

What it is: Vector Search, the ultra high performance vector database powering Vertex AI Search, DIY RAG, and other embedding use cases at global scale, now offers hybrid search in Public Preview. 

Why it matters: Embeddings are numerical representations that capture semantic relationships across complex data (text, images, etc.). Embeddings power multiple use cases, including recommendation systems, ad serving, and semantic search for RAG. Hybrid search combines vector-based and keyword-based search techniques to ensure the most relevant and accurate responses for users. 

Get started: Visit documentation to learn more about Vector Search.

LangChain on Vertex

What it is: An agent development SDK and container runtime for LangChain. With LangChain on Vertex AI you can select the model you want to work with, define tools to access external APIs, structure the interface between the user and the system components in an orchestration framework, and deploy the framework to a managed runtime.

Why it matters: LangChain on Vertex AI simplifies and speeds up deployment while being secure, private and scalable. 

Get started: Visit documentation to learn more. 

Vertex AI extensions, function calling and ​​data connectors 

What it is: 

Vertex AI extensions are pre-built reusable modules to connect a foundation model to a specific API or tool. For example, our new code interpreter extension enables models to execute tasks that entail running Python code, such as data analysis, data visualization, and mathematical operations. 

Vertex AI function calling enables a user to describe a set of functions or APIs and have Gemini models intelligently select, for a given query, the right API or function to call, along with the appropriate API parameters.

Vertex AI data connectors help ingest data from enterprise and third-party applications like ServiceNow, Hadoop, and Salesforce, connecting generative applications to commonly-used enterprise systems.

Why it matters: With these capabilities, Vertex AI Agent Builder makes it easy to augment grounding outputs and take action on your user’s behalf. 

Get started: Visit documentation to learn more about Vertex AI extensions, function calling and ​​data connectors.

Firebase Genkit 

What it is: Genkit is an open-source TypeScript/JavaScript and Go framework designed by Firebase to simplify the development, deployment, and monitoring of production-ready AI applications.

Why it matters: With the Vertex AI plugin for Genkit, developers can now take advantage of Google models like Gemini and Imagen 2, as well as text embeddings. Additionally Vertex Eval Service is baked into the Genkit local development experience along with OpenTelemetry tracing.

Get started: Learn more in documentation.

LlamaIndex on Vertex AI

What it is: LlamaIndex on Vertex AI simplifies building your own search engine for retrieval-augmented generation (RAG), from data ingestion and transformation to embedding, indexing, retrieval, and generation.

Why it matters: Vertex AI customers can leverage Google’s models and AI-optimized infrastructure alongside LlamaIndex’s simple, flexible, open-source data framework, to connect custom data sources to generative models. 

Get started: Visit documentation to learn more.

Built on a foundation of scale & enterprise readiness
The revolutionary nature of generative AI requires a platform that offers privacy, security, control, and compliance capabilities organizations can rely on. Google Cloud is committed to helping our customers leverage the full potential of generative AI with privacy, security, and compliance capabilities. Our goal is to build trust by protecting systems, enabling transparency, and offering flexible, always-available infrastructure, all while grounding efforts in our AI principles.
Dynamic Shared Quota

What it is: With Dynamic Shared Quota, we offer increasing the quota limits for a model (online serving) to the maximum allowed per region. This way we limit the number of queries per second (QPS) that customers can run by the shared capacity of all the queries running on a Servo station (multi-region), instead of limiting a customer’s QPS by a quota. Dynamic Shared Quota is only applicable to Pay-as-you-go Online Serving. For customers that require a consistent or more predictable service level, including SLAs, we offer Provision Throughput.

Why it matters: By dynamically distributing on-demand capacity among all queries being processed for Pay-as-you-go customers, Google Cloud has eliminated the need to submit quota increase requests (QIRs). Customers can still set a self-imposed quota called a consumer quota override to control cost and prevent budget overruns.

Get started: Learn more in documentation.

Provisioned Throughput is GA 

What it is: Provisioned throughput lets customers responsibly scale their usage of Google’s first-party models, like 1.5 Flash, providing assurances for both capacity and price. 

Why it matters: This Vertex AI feature brings predictability and reliability to customer production workloads, giving them the assurance required to scale gen AI workloads aggressively.  We have also made it easier than ever for customers to set up PT via a Self Service flow. Customers can now estimate their needs and purchase Provisioned Throughput for Google’s 1P foundation models via the console, bringing the E2E experience down from weeks to minutes for pre-approved orders subject to available capacity and removing the need for manual order forms.

Get started: Follow these steps to purchase a Provisioned Throughput subscription.

Data residency for data stored at-rest guarantees in more countries

What it is: We have data residency for data stored at-rest guarantees in 23 countries (13 of which were added in 2024), with additional guarantees around limiting related ML processing to the US and EU. We are also working on expanding our ML processing commitments to eight more countries, starting with four countries in 2024.

Why it matters: Customers, especially those from regulated industries, demand control over where their data is stored and processed when using generative AI capabilities. 

Get started: Learn more here.

To keep up with all of the latest releases, don’t forget to check our Vertex AI release notes. 
All of these enhancements are a direct response to what you, our customers, have been asking for. We believe an enterprise AI platform is key to success in production and our goal is to not just build the best platform, but to provide an AI ecosystem that makes enterprise-scale AI accessible.
To learn about how Vertex AI can help you, contact us for a free consultation.

aside_block
<ListValue: [StructValue([('title', 'Talk to an expert'), ('body', <wagtail.rich_text.RichText object at 0x3ebc3d11b880>), ('btn_text', 'Contact sales'), ('href', 'https://inthecloud.withgoogle.com/global-gen-ai-contact-sales/cs-bsm.html?utm_source=cgc-blog&utm_medium=blog&utm_campaign=FY23-Q3-global-ENDM228-website-sm-gen-ai-contact-us-lp&utm_content=vertex-ai-evergreen-announcements&utm_term=-'), ('image', None)])]>

Quelle: Google Cloud Platform

Enhancing LLM quality and interpretability with the Vertex AI Gen AI Evaluation Service

Developers harnessing the power of large language models (LLMs) often encounter two key hurdles: managing the inherent randomness of their output and addressing their occasional tendency to generate factually incorrect information. Somewhat like rolling dice, LLMs offer a touch of unpredictability, generating different responses even when given the same prompt. While this randomness can fuel creativity, it can also be a stumbling block when consistency or factual accuracy is crucial. Moreover, the occasional “hallucinations” – where the LLM confidently presents misinformation – can undermine trust in its capabilities. The challenge intensifies when we consider that many real-world tasks lack a single, definitive answer. Whether it’s summarizing complex information, crafting compelling marketing copy, brainstorming innovative product ideas, or drafting persuasive emails, there’s often room for multiple valid solutions.In this blog post and accompanying notebook, we’ll explore how to tackle these challenges by introducing a new workflow which works by generating a diverse set of LLM-generated responses and employing the Vertex Gen AI Evaluation Service to automate the selection process of the best response and provide associated quality metrics and explanation. This process is also extensible to multimodal input and output and stands to benefit almost all use cases across industries and LLMs.Picture this: a financial institution striving to summarize customer conversations with banking advisors. The hurdle? Ensuring these summaries are grounded in reality, helpful, concise, and well-written. With numerous ways to craft a summary, the quality varied greatly. Here is how they leveraged the probabilistic nature of LLMs and the Vertex Gen AI Evaluation Service to elevate the performance of the LLM-generated summaries.Step 1: Generate Diverse ResponsesThe core idea here was to think beyond the first response. Causal decoder-based LLMs have a touch of randomness built in, meaning they sample each word probabilistically. So, by generating multiple, slightly different responses, we boost the odds of finding a perfect fit. It’s like exploring multiple paths, knowing that even if one leads to a dead end, another might reveal a hidden gem.For example, imagine asking an LLM, “What is the capital of Japan?” You might get a mix of responses like “Kyoto was the capital city of Japan,” “Tokyo is the current capital of Japan,” or even “Tokyo was the capital of Japan.” By generating multiple options, we increase our chances of getting the most accurate and relevant answer.To put this into action, the financial institution used an LLM to generate five different summaries for each transcript. They adjusted the LLM’s “temperature,” which controls the randomness of output, to a range of 0.2 to 0.4, to encourage just the right amount of diversity without straying too far from the topic. This ensured a range of options, increasing the likelihood of finding an ideal, high-quality summary.Step 2: Find the Best ResponseNext came the need to search through the set of diverse responses and pinpoint the best one. To do this automatically, the financial institution applied the pairwise evaluation approach available in the Vertex Gen AI Evaluation Service. Think of it as a head-to-head showdown between responses. We pit response pairs against each other, judging them based on the original instructions and context to identify the response that aligns most closely with the user’s intent.Continuing the example above to illustrate, let’s say we have those three responses about Japan’s capital. We want to find the best one using pairwise comparisons:Response 1 vs Response 2: The API favors Response 2, potentially explaining, “While Response 1 is technically correct, it doesn’t directly answer the question about the current capital of Japan.”Response 2 (best response so far) vs Response 3: Response 2 wins again! Response 3 stumbles by using the past tense.After these two rounds of comparison, we conclude that Response 2 is the best answer.In the financial institution’s case, they compared their five generated summaries in pairs to select the best one.Step 3: Assess if the Response is Good EnoughThe workflow then takes the top-performing response (Response 2) from the previous step and uses the pointwise evaluation service to assess it. This evaluation assigns quality scores and generates human-readable explanations for those scores across various dimensions, such as accuracy, groundedness, and helpfulness. This process not only highlights the best response but also provides insights into why the model generated this response, and also why it’s considered superior to the other responses, fostering trust and transparency in the system’s decision-making. In the case of the financial institution, they now used the summarization-related metrics in pointwise evaluation on the winning response to obtaining an explanation of how this answer is grounded, helpful, and high-quality. We can choose to return just the best response or include its associated quality metrics and explanation for greater transparency.In essence, the workflow (as illustrated in this blog’s banner) encompasses generating a variety of LLM responses, systematically evaluating them, and selecting the most suitable one—all while providing insights into why that particular response is deemed optimal. Get started by exploring our sample notebook and adapting it to fit with your use case. You can reverse the order of pairwise and pointwise evaluations, by ranking individual responses based on their pointwise scores and then conducting pairwise comparisons only on the top candidates. Further, while this example focuses on text, this approach can be applied to any modality or any use case including but not limited to question answering and summarization like illustrated in this blog. Finally, if you need to minimize latency, both workflows can benefit greatly from parallelizing the various API calls.Take the next stepBy embracing the inherent variability of LLMs and utilizing the Vertex Gen AI Evaluation Service, we can transform challenges into opportunities. Generating diverse responses, systematically evaluating them, and selecting the best option with clear explanations empowers us to unlock the full potential of LLMs. This approach not only enhances the quality and reliability of LLM outputs but also fosters trust and transparency. Start exploring this approach in our sample notebook and check out the documentation for the Vertex Gen AI Evaluation Service.
Quelle: Google Cloud Platform

What’s new with Google Cloud – 2023

Want to know the latest from Google Cloud? Find it here in one handy location. Check back regularly for our newest updates, announcements, resources, events, learning opportunities, and more. Tip: Not sure where to find what you’re looking for on the Google Cloud blog? Start here: Google Cloud blog 101: Full list of topics, links, and resources.Week of Dec 10 – Dec 15Announcing Launch of Cross Cloud Materialized Views: To help customers on their cross-cloud analytics journey, today we are thrilled to announce the public preview of BigQuery Omni cross-cloud materialized views (aka cross-cloud MVs). Cross-cloud MVs allow customers to very easily create a summary materialized view on GCP from base data assets available on another cloud. Cross-cloud MVs are automatically and incrementally maintained as base tables change, meaning only a minimal data transfer is necessary to keep the materialized view on GCP in sync. The result is an industry-first capability that empowers customers to perform frictionless, efficient, and cost-effective cross-cloud analyticsWeek of Nov 27 – Dec 1Google Cloud Global Cloud Service Provider of the Year. Google Cloud is thrilled to be recognized as Palo Alto Networks 2023 Global Cloud Service Provider of the Year and Global Cortex Partner of the Year. Google Cloud and Palo Alto Networks are dedicated to working together to support customer cloud journeys with an array of jointly engineered and integrated security solutions that enable digital innovation with ease. Read the Palo Alto Networks blog.GKE Enterprise edition free trial: We have announced the general availability of GKE Enterprise, the premium edition of Google Kubernetes Engine (GKE) recently. With GKE Enterprise, companies can increase velocity across multiple teams, easily and securely run their most important business-critical apps and the latest AI/ML workloads safely at scale with a fully integrated and managed solution from Google Cloud. Start the 90-day free trial today with the GKE Enterprise edition by going to the GKE console and clicking on the “Learn about GKE Enterprise” button.Assured Workloads Resource ValidationIn our new blog post on Cost Management in BigQuery, you’ll learn how to use budgets and custom quota to help you stay on top of your spending and prevent surprises on your cloud bill. The interactive tutorials linked in the article will help you set them up for your own Google Cloud projects in no time!Leverage the transformative power of generative AI to elevate your customer service. Discover how you can optimize ROI, enhance customer satisfaction, and revolutionize your contact center operations with Google’s latest conversational AI offerings in this new blog.Week of Nov 13 – Nov 17In the first of our new Sketchnote series on Cloud FinOps, Erik and Pathik dive into what Cloud FinOps is, and how it can help your organization optimize its cloud budget.Week of Oct 30 – Nov 3Join Google Cloud’s product management leadership for a Data Analytics Innovation Roadmap session on November 13th. In this session, we will go through recent innovations, strategy and plans for BigQuery, Streaming Analytics, Data Lakes, Data Integration, and GenAI. This session will give you insight into Google’s feature development and will help your team plan your data analytics strategy.Hear from Google Cloud experts on modernizing software delivery with generative AI, running AI/ML workloads on GKE, the future of AI-infused apps, and more at Digital Transform: the future of AI-powered apps, November 15th.Vertex AI Search: Read about exciting new generative AI features coming to Vertex AI Search our platform to create search based applications for your business. Vertex AI Search provides customers with a tunable Retrieval Augmented Generation (RAG) system for information discovery. Learn more in this blog.Vector similarity search: If you are looking to build an an e ecommerce recommendations engine or ad serving or other DIY application based on ANN aka vector similarity search dive into our vector search capability which is a part of the Vertex AI Search platform. We’ve expanded features and made it easier then ever for developers to get started building their apps.Cloud Deploy – Deploy hooks (GA) allow users to specify and execute pre- and post- deploy actions using Cloud Deploy. This allows customers to run infrastructure deployment, database schema updates, and other activities immediately before a deploy job, and cleanup operations as part of a post (successful) deploy job. Learn MoreCloud Deploy – Cloud Deploy now uses Skaffold 2.8 as the default Skaffold version for all target types. Learn MoreArtifact Registry – Artifact Registry remote repositories are now generally available (GA). Remote repositories store artifacts from external sources such as Docker Hub or PyPI. A remote repository acts as a proxy for the external source so that you have more control over your external dependencies. Learn MoreArtifact Registry – Artifact Registry virtual repositories are now generally available (GA). Virtual repositories act as a single access point to download, install, or deploy artifacts in the same format from one or more upstream repositories. Learn MoreWeek of Oct 2 – Oct 6ABAP SDK for Google Cloud now supports 40+ more APIs, an additional authentication mechanism and enhanced developer productivity for SAP ABAP developers. Learn more in this blog post.Week of Sep 25 – Sep 29Our newly published Storage decision tree helps you research and select the storage services in Google Cloud that best match your specific workload needs and the accompanying blog provides an overview of the services offered for block storage, object storage, NFS and Multi-Writer file storage, SMB storage, and storage for data lakes and data warehouses.Week of Sep 18- Sep 22Meet the inaugural cohort of the Google for Startups Accelerator: AI First program featuring groundbreaking businesses from eight countries across Europe and Israel using AI and ML to solve complex problems. Learn how Google Cloud empowers these startups and check out the selected ventures here.BigQuery is introducing new SQL capabilities for improved analytics flexibility, data quality and security. Some examples include schema support for Flexible column name, Authorized store proceduces, ANY_VALUE (HAVING) also known as MAX_BY and MIN_BY and many more. Check out full details here.Cloud Logging is introducing to Preview the ability to save charts from Cloud Logging’s Log Analytics to a custom dashboard in Cloud Monitoring. Viewing, copying and sharing the dashboards are supported in Preview. For more information, see Save a chart to a custom dashboard.Cloud Logging now supports customizable dashboards in its Logs Dashboard. Now you, can add your own charts to see what’s most valuable to you on the Logs Dashboard. Learn more here.Cloud Logging launches several usability features for effective troubleshooting. Learn more in this blog post.Search your logs by service name with the new option in Cloud Logging. Now you can use the Log fields to select by service which makes it easier to quickly find your Kubernetes container logs. Check out the details here.Community Security Analytics (CSA) can now be deployed via Dataform to help you analyze your Google Cloud security logs. Dataform simplifies deploying and operating CSA on BigQuery, with significant performance gains and cost savings. Learn more why and how to deploy CSA with Dataform in this blog post.Dataplex data profiling and AutoDQ are powerful new features that can help organizations to improve their data quality and build more accurate and reliable insights and models. These features and now Generally Available. Read more in this blog post.Week of Sep 4 – Sep 8Introducing Looker’s Machine Learning Accelerator. This easy to install extension allows business users to train, evaluate, and predict with machine learning models right in the Looker interface.Learn about how Freestar has built a super low latency, globally distributed application powered by Memorystore and the Envoy proxy. This reference walks users through the finer details of architecture and configuration, that they can easily replicate for their own needs.Week of Aug 28 – Sep 1You can access comprehensive and up-to-date environmental information to develop sustainability solutions and help people adapt to the impacts of climate change through Google Maps Platform’s environment APIs. The Air Quality, and Solar APIs are generally available today. Get started or learn more in this blog post.Google Cloud’s Global Partner Ecosystems & Channels team launched the Industry Value Networks (IVN) initiative at Google Cloud NEXT ’23. IVNs combine expertise and offerings from systems integrators (SIs), independent software vendors (ISVs) and content partners to create comprehensive, differentiated, repeatable, and high-value solutions that accelerate time-to-value and reduce risk for customers. To learn more about the IVN initiative, please see this blog postWeek of Aug 21 – Aug 25You can now easily export data from Earth Engine into BigQuery with our new connector. This feature allows for improved workflows and new analyses that combine geospatial raster and tabular data. This is the first step in toward deeper interoperability between the two platforms, supporting innovations in geospatial sustainability analytics. Learn more in this blog post or join our session at Cloud Next.Week of Aug 14 – Aug 18You can now view your log query results as a chart in the Log Analytics page in Cloud Logging. With this new capability available in Preview, users can write a SQL filter and then use the charting configuration to build a chart. For more information, see Chart query results with Log Analytics.Week of Aug 7 – Aug 11You can now use Network Analyzer and Recommender API to query the IP address utilization of your GCP subnets, to identify subnets that might be full or oversized. Learn more in a dedicated blog post here.Memorystore has introduced version support for Redis 7.0. Learn more about the included features and upgrade your instance today!Week of July 31 – Aug 4Attack Path Simulation is now generally available in Security Command Center Premium. This new threat prevention capability automatically analyzes a customer’s Google Cloud environment to discover attack pathways and generate attack exposure scores to prioritize security findings. Learn more. Get started now.Week of July 24-28Cloud Deploy has updated the UI with the ability to Create a Pipeline along with a Release. The feature is now GA. Read moreOur newly published Data & Analytics decision tree helps you select the services on Google Cloud that best match your data workloads needs, and the accompanying blog provides an overview of the services offered for data ingestion, processing, storage, governance, and orchestration.Customer expectations from the ecommerce platforms are at all time high and they now demand a seamless shopping experience across platforms, channels and devices. Establishing a secure and user-friendly login platform can make it easier for users to self-identify and help retailers gain valuable insights into customer’s buying habits. Learn more about how they can better manage customer identities to support an engaging ecommerce user experience using Google Cloud Identity Platform.Our latest Cloud Economics post just dropped, exploring how customers can benchmark their IT spending against peers to optimize investments. Comparing metrics like tech spend as a percentage of revenue and OpEx uncovers opportunities to increase efficiency and business impact. This data-driven approach is especially powerful for customers undergoing transformation.Week of July 17-21Cloud Deploy now supports deploy parameters. With deploy parameters you can pass parameters for your release, and those values are provided to the manifest or manifests before those manifests are applied to their respective target. A typical use for this would be to apply different values to manifests for different targets in a parallel deployment. Read moreCloud Deploy is now listed among other Google Cloud services which can be configured to meet Data Residency Requirements. Read moreLog Analytics in Cloud Logging now supports most regions. Users can now upgrade buckets to use Log Analytics in Singapore, Montréal, London, Tel Aviv and Mumbai. Read more for the full list of support regions.Week of July 10-14Cloud CDN now supports private origin authentication in GA. This capability improves security by allowing only trusted connections to access the content on your private origins and preventing users from directly accessing it.Workload Manager – Guided Deployment Automation is now available in Public Preview, with initial support for SAP solutions. Learn how to configure and deploy SAP workloads directly from a guided user interface, leveraging end-to-end automation built on Terraform and Ansible.Artifact Registry – Artifact registry now supports clean up policies now in Preview. Cleanup policies help you manage artifacts by automatically deleting artifacts that you no longer need, while keeping artifacts that you want to store. Read moreWeek of July 3-7Cloud Run jobs now supports long-running jobs. A single Cloud Run jobs task can now run for up to 24 hours. Read More.How Google Cloud NAT helped strengthen Macy’s security. Read moreWeek of June 26-30Cloud Deploy parallel deployment is now generally available. You can deploy to a target that’s configured to represent multiple targets, and your application is deployed to those targets concurrently. Read More.Cloud Deploy canary deployment strategy is now generally available. A canary deployment is a progressive rollout of an application that splits traffic between an already-deployed version and a new version. Read MoreWeek of June 19-23Google Cloud’s Managed Service for Prometheus now supports Prometheus exemplars. Exemplars provide cross-signals correlation between your metrics and your traces so you can more easily pinpoint root cause issues surfaced in your monitoring operations.Managing logs across your organization is now easier with the general availability of user-managed service accounts. You can now choose your own service account when sending logs to a log bucket in a different project.Data Engineering and Analytics Day – Join Google Cloud experts on June 29th to learn about the latest data engineering trends and innovations, participate in hands-on labs, and learn best practices of Google Cloud’s data analytics tools. You will gain a deeper understanding of how to centralize, govern, secure, streamline, analyze, and use data for advanced use cases like ML processing and generative AI.Week of June 5-9TMI: Shifting Down, Not Left- The first post in our new modernization series, The Modernization Imperative. Here, Richard Seroter talks about the strategy of ‘shifting down’ and relying on managed services to relieve burdens on developers.Cloud Econ 101: The first in a new series on optimizing cloud tools to achieve greater return on your cloud investments. Join us biweekly as we explore ways to streamline workloads, and explore successful cases of aligning technology goals to drive business value.The Public Preview: of Frontend Mutual TLS Support on Global External HTTPS Load Balancing is now available. Now you can use Global External HTTPS Load Balancing to offload Mutual TLS authentication for your workloads. This includes client mTLS for Apigee X Northbound Traffic using Global HTTPS Load Balancer.FinOps from the field: How to build a FinOps Roadmap – In a world where cloud services have become increasingly complex, how do you take advantage of the features, but without the nasty bill shock at the end? Learn how to build your own FinOps roadmap step by step, with helpful tips and tricks from FinOps workshops Google has completed with customers.We are now offering up to $1M of financial protection to help cover the costs of undetected cryptomining attacks. This is a new program only for Security Command Center Premium customers. Security Command Center makes Google Cloud a safe place for your applications and data. Read about this new program in our blog.Global External HTTP(S) Load Balancer and Cloud CDN’s advanced traffic management using flexible pattern matching is now GA. This allows you to use wildcards anywhere in your path matcher. You can use this to customize origin routing for different types of traffic, request and response behaviors, and caching policies. In addition, you can now use results from your pattern matching to rewrite the path that is sent to the origin.Security Command Center (SCC) Premium, our built-in security and risk management solution for Google Cloud, is now generally available for self-service activation for full customer organizations. Customers can get started with SCC in just a few clicks in the Google Cloud console. There is no commitment requirement, and pricing is based on a flexible pay-as-you-go model.Dataform is Generally Available. Dataform offers an end-to-end experience to develop, version control, and deploy SQL pipelines in BigQuery. Using a single web interface, data engineers and data analysts of all skill levels can build production-grade SQL pipelines in BigQuery while following software engineering best practices such as version control with Git, CI/CD, and code lifecycle management. Learn more.Week of May 29 – June 2Google Cloud Deploy. The price of an active delivery pipeline is reduced. Also, single-target delivery pipelines no longer incur a charge. Underlying service charges continue to apply. See Pricing Page for more details.Week of May 22-26Security Command Center (SCC) Premium pricing for project-level activation is now 25% lower for customers who use SCC to secure Compute Engine, GKE-Autopilot, App Engine and Cloud SQL. Please see our updated rate card. Also, we have expanded the number of finding types available for project-level Premium activations to help make your environment more secure. Learn more.Vertex AI Embeddings for Text: Grounding LLMs made easy: Many people are now starting to think about how to bring Gen AI and large language models (LLMs) to production services. You may be wondering “How to integrate LLMs or AI chatbots with existing IT systems, databases and business data?”, “We have thousands of products. How can I let LLM memorize them all precisely?”, or “How to handle the hallucination issues in AI chatbots to build a reliable service?”. Here is a quick solution: grounding with embeddings and vector search. What is grounding? What are embedding and vector search? In this post, we will learn these crucial concepts to build reliable Gen AI services for enterprise use with live demos and source code.Week of May 15-19Introducing the date/time selector in Log Analytics in Cloud Logging. You can now easily customize the date and time range of your queries in the Log Analytics page by using the same date/time-range selector used in Logs Explorer, Metrics Explorer and other Cloud Ops products. There are several time range options, such as preset times, custom start and end times, and relative time ranges. For more information, see Filter by time in the Log Analytics docs.Cloud Workstations is now GA. We are thrilled to announce the general availability of Cloud Workstations with a list of new enhanced features, providing fully managed integrated development environments (IDEs) on Google Cloud. Cloud Workstations enables faster developer onboarding and increased developer productivity while helping support your compliance requirements with an enhanced security posture. Learn MoreWeek of May 8 – 14Introducing BigQuery differential privacy, SQL building blocks that analysts and data scientists can use to anonymize their data. We are also partnering with Tumult Labs to help Google Cloud customers with their differential privacy implementations.Scalable electronic trading on Google Cloud: A business case with BidFX: Working with Google Cloud, BidFX has been able to develop and deploy a new product called Liquidity Provision Analytics (“LPA”), launching to production within roughly six months, to solve the transaction cost analysis challenge in an innovative way. LPA will be offering features such as skew detection for liquidity providers, execution time optimization, pricing comparison, top of book analysis and feedback to counterparties. Read more here.AWS EC2 VMs discovery and assessment – mFit can discover EC2 VMs inventory in your AWS region and collect guest level information from multiple VMs to provide technical fit assessment for modernization. See demo video.Generate assessment report in Microsoft Excel file – mFit can generate detailed assessment report in Microsoft Excel (XLSX) format which can handle large amounts of VMs in a single report (few 1000’s) which an HTML report might not be able to handle.Regulatory Reporting Platform: Regulatory reporting remains a challenge for financial services firms. We share our point of view on the main challenges and opportunities in our latest blog, accompanied by an infographic and a customer case study from ANZ Bank. We also wrote a white paper for anyone looking for a deeper dive into our Regulatory Reporting Platform.Google is partnering with regional carriers Chunghwa Telecom, Innove (subsidiary of Globe Group) and AT&T to deliver the TPU (Taiwan-Philippines-U.S.) cable system — connecting Taiwan, Philippines, Guam, and California — to support growing demand in the APAC region. We are committed to providing Google Cloud customers with a resilient, high-performing global network. NEC is the supplier, and the system is expected to be ready for service in 2025.Week of May 1 – 5Microservices observability is now generally available for C++, Go and Java. This release includes a number of new features and improvements, making it easier than ever to monitor and troubleshoot your microservices applications. Learn more on our user guide.Google Cloud Deploy Google Cloud Deploy now supports Skaffold 2.3 as the default Skaffold version for all target types. Release Notes.Cloud Build You can now configure Cloud Build to continue executing a build even if specified steps fail. This feature is generally available. Learn more hereWeek of April 24-28General Availability: Custom Modules for Security Health Analytics is now generally available. Author custom detective controls in Security Command Center using the new custom module capability.Next generation Confidential VM is now available in Private Preview with a Confidential Computing technology called AMD Secure Encrypted Virtualization-Secure Nested Paging (AMD SEV-SNP) on general purpose N2D machines. Confidential VMs with AMD SEV-SNP enabled builds upon memory encryption and adds new hardware-based security protections such as strong memory integrity, encrypted register state (thanks to AMD SEV-Encrypted State, SEV-ES), and hardware-rooted remote attestation. Sign up here!Selecting Tier_1 networking for your Compute Engine VM can give you the bandwidth you need for demanding workloads.Check out this blog on Increasing bandwidth to Compute Engine VMs with TIER_1 networking.Week of April 17-21Use Terraform to manage Log Analytics in Cloud Logging You can now configure Log Analytics on Cloud Logging buckets and BigQuery linked datasets by using the following Terraform modules:Google_logging_project_bucket_configgoogle_logging_linked_datasetWeek of April 10-14Assured Open Source Software is generally available for Java and Python ecosystems. Assured OSS is offered at no charge and provides an opportunity for any organization that utilizes open source software to take advantage of Google’s expertise in securing open source dependencies.BigQuery change data capture (CDC) is now in public preview! BigQuery CDC provides a fully-managed method of processing and applying streamed UPSERT and DELETE operations directly into BigQuery tables in real time through the BigQuery Storage Write API. This further enables the real-time replication of more classically transactional systems into BigQuery, which empowers cross functional analytics between OLTP and OLAP systems. Learn more here.Week of April 3-7New Visualization tools for Compute Engine Fleets TheObservability tab in the Compute Engine console VM List page has reached General Availability. The new Observability tab is an easy way to monitor and troubleshoot the health of your fleet of VMsDatastream for BigQuery is Generally Available! Datastream for BigQuery is generally available, offering a unique, truly seamless and easy-to-use experience that enables near-real time insights in BigQuery with just a few steps. Using BigQuery’s newly developed change data capture (CDC) and Storage Write API’s UPSERT functionality, Datastream efficiently replicates updates directly from source systems into BigQuery tables in real-time. You no longer have to waste valuable resources building and managing complex data pipelines, self-managed staging tables, tricky DML merge logic, or manual conversion from database-specific data types into BigQuery data types. Just configure your source database, connection type, and destination in BigQuery and you’re all set. Datastream for BigQuery will backfill historical data and continuously replicate new changes as they happen.Now available! Build an analytics lakehouse on Google Cloud whitepaper. The analytics lakehouse combines the benefits of data lakes and data warehouses without the overhead of each. In this paper, we discuss the end-to-end architecture which enable organizations to extract data in real-time regardless of which cloud or datastore the data reside in, use the data in aggregate for greater insight and artificial intelligence (AI) – all with governance and unified access across teams. Download now.Now Available! Google Cloud Deploy now supports canary release as a deployment strategy. This feature is supported in Preview. Learn moreGeneral Availability : Cloud Run services as backends to Internal HTTP(S)Load Balancers and Regional External HTTP(S)Load Balancers. Internal load balancers allow you to establish private connectivity between Cloud Run services and other services and clients on Google Cloud, on-premises, or on other clouds. In addition you get custom domains, tools to migrate traffic from legacy services, Identity-aware proxy support, and more. Regional external load balancer, as the name suggests, is designed to reside in a single region and connect with workloads only in the same region, thus helps you meet your regionalization requirements. Learn more.Week of March 27-31Last chance: Register to attend Google Data Cloud & AI Summit Join us on Wednesday, March 29, at 9 AM PDT/12 PM EDT to discover how you can use data and AI to reveal opportunities to transform your business and make your data work smarter. Find out how organizations are using Google Cloud data and AI solutions to transform customer experiences, boost revenue, and reduce costs. Register today for this no cost digital event.New BigQuery editions: flexibility and predictability for your data cloudAt the Data Cloud & AI Summit, we announced BigQuery pricing editions—Standard, Enterprise and Enterprise Plus—that allow you to choose the right price-performance for individual workloads. Along with editions, we also announced autoscaling capabilities that ensure you only pay for the compute capacity you use, and a new compressed storage billing model that is designed to reduce your storage costs. Learn more about latest BigQuery innovations and register for the upcoming BigQuery roadmap session on April 5, 2023.Introducing Looker Modeler: A single source of truth for BI metricsAt the Data Cloud & AI Summit, we introduced a standalone metrics layer we call Looker Modeler, available in preview in Q2. With Looker Modeler, organizations can benefit from consistent governed metrics that define data relationships and progress against business priorities, and consume them in BI tools such as Connected Sheets, Looker Studio, Looker Studio Pro, Microsoft Power BI, Tableau, and ThoughtSpot.Cloud Workstations is now available in more regions. Cloud Workstations is now available in asia-south1 (India), us-east4 (Virginia, North America), europe-west6 (Switzerland), and europe-west9 (France). The full list of regions is here.Bucket based log based metrics—now generally available—allow you to track, visualize and alert on important logs in your cloud environment from many different projects or across the entire organization based on what logs are stored in a log bucket.NEW Customer Blog! Faced with strong data growth, Squarespace made the decision to move away from on-premises Hadoop to a cloud-managed solution for its data platform. Learn how they reduced the number of escalations by 87% with the analytics lakehouse on Google Cloud. Read nowWeek of March 20-24Chronicle Security Operations Feature RoundupBringing a modern and unified security operations experience to our customers is and has been a top priority with the Google Chronicle team. We’re happy to show continuing innovation and even more valuable functionality. In our latest release roundup we’ll highlight a host of new capabilities focused on delivering improved context, collaboration, and speed to handle alerts faster and more effectively. Learn how our newest capabilities enable security teams to do more with less here.Announcing Google’s Data Cloud & AI Summit, March 29th!Can your data work smarter? How can you use AI to unlock new opportunities? Join us on Wednesday, March 29, to gain expert insights, new solutions, and strategies to reveal opportunities hiding in your company’s data. Find out how organizations are using Google Cloud data and AI solutions to transform customer experiences, boost revenue, and reduce costs. Register today for this no cost digital event.Artifact Registry Feature Preview – Artifact Registry now supports immutable tags for Docker repositories. If you enable this setting, an image tag always points to the same image digest, including the default latest tag. This feature is in Preview. Learn moreWeek of March 13 – 17Building the most open and innovative AI ecosystemIn addition to the news this week on AI products, Google Cloud has also announced new partnerships, programs, and resources. This includes bringing bringing the best of Google’s infrastructure, AI products, and foundation models to partners at every layer of the AI stack: chipmakers, companies building foundation models and AI platforms, technology partners enabling companies to develop and deploy machine learning (ML) models, app-builders solving customer use-cases with generative AI, and global services and consulting firms that help enterprise customers implement all of this technology at scale. Learn more.From Microbrows to MicroservicesUlta Beauty is building their digital store of the future, but to maintain control over their new modernized application they turned to Anthos and GKE – Google Cloud’s managed container services, to provide an eCommerce experience as beautiful as their guests. Read our blog to see how a newly-minted Cloud Architect learnt Kubernetes and Google Cloud to provide the best possible architecture for his developers. Learn more.To prepare for the busiest shopping season of the year, Black Friday and Cyber Monday, Lowe’s relies heavily on Google’s agile SRE Framework to ensure business and technical alignment, manage bots, and create an always-available shopping experience. Read more.Now generally available, understand and trust your data with Dataplex data lineage – a fully managed Dataplex capability that helps you understand how data is sourced and transformed within the organization. Dataplex data lineage automatically tracks data movement across BigQuery, BigLake, Cloud Data Fusion (Preview), and Cloud Composer (Preview), eliminating operational hassles around manual curation of lineage metadata. Learn more here.Rapidly expand the reach of Spanner databases with read-only replicas and zero-downtime moves. Configurable read-only replicas let you add read-only replicas to any Spanner instance to deliver low latency reads to clients in any geography. Alongside Spanner’s zero-downtime instance move service, you have the freedom to move your production Spanner instances from any configuration to another on the fly, with zero downtime, whether it’s regional, multi-regional, or a custom configuration with configurable read-only replicas. Learn more here.Week of March 6 – 10Automatically blocking project SSH keys in Dataflow is now GA. This service option allows Dataflow users to prevent their Dataflow worker VMs from accepting SSH keys that are stored in project metadata, and results in improved security. Getting started is easy: enable the block-project-ssh-keys service option while submitting your Dataflow job.Celebrate International Women’s Day Learn about the leaders driving impact at Google Cloud and creating pathways for other women in their industries. Read more.Google Cloud Deploy now supports Parallel Deployment to GKE and Cloud Run workloads. This feature is in Preview. Read more.Sumitovant doubles medical research output in one year using LookerSumitovant is a leading biopharma research company that has doubled their research output in one year alone. By leveraging modern cloud data technologies, Sumitovant supports their globally distributed workforce of scientists to develop next generation therapies using Google Cloud’s Looker for trusted self-service data research. To learn more about Looker check out https://cloud.google.com/lookerWeek of Feb 27 – March 3Accelerate Queries on your BigLake Tables with Cached Metadata (Preview!)Make your queries on BigLake Tables go faster by enabling metadata caching. Your queries will avoid expensive LIST operation for discovering files in the table and experience faster file and hive partition pruning. Follow the documentation here.Google Cloud Deploy support for deployment verification is now GA! Read more or Try the DemoAdd geospatial intelligence to your Retail use cases by leveraging the CARTO platform on top of your data in BigQueryLocation data will add a new dimension to your Retail use cases, like site selection, geomarketing, and logistics and supply chain optimization. Read more about the solution and various customer implementations in the CARTO for Retail Reference Guide, and see a demonstration in this blog.Week of Feb 20 – Feb 24Start your digital transformation by embarking on a hybrid cloud journey with Anthos. Anthos helps you modernize your application and infrastructure in place and build a unified Kubernetes fabric between your on prem environments and the Google cloud. The newly published Anthos hybrid cloud architecture reference design guide provides opinionated guidance to deploy Anthos in a hybrid environment to address some common challenges that you might encounter. Check out the architecture reference design guidehere to accelerate your journey to hybrid cloud and containerization.Logs for Network Load Balancingand logs for Internal TCP/UDP Load Balancingare now GA! Logs are aggregated per-connection and exported in near real-time, providing useful information, such as 5-tuples of the connection, received bytes, and sent bytes, for troubleshooting and monitoring the pass-through Google Cloud Load Balancers. Further, customers can include additional optional fields, such as annotations for client-side and server-side GCE and GKE resources, to obtain richer telemetry.Week of Feb 13 – Feb 17,Announcing Google’s Data Cloud & AI Summit, March 29th!Can your data work smarter? How can you use AI to unlock new opportunities? Register for Google Data Cloud & AI Summit, a digital event for data and IT leaders, data professionals, developers, and more to explore the latest breakthroughs. Join us on Wednesday, March 29, to gain expert insights, new solutions, and strategies to reveal opportunities hiding in your company’s data. Find out how organizations are using Google Cloud data and AI solutions to transform customer experiences, boost revenue, and reduce costs. Register today for this no cost digital event.Leverege uses BigQuery as a key component of its data and analytics pipeline to deliver innovative IoT solutions at scale. As part of the Built with BigQuery program, this blog post goes into detail about Leverege IoT Stack that runs on Google Cloud to power business-critical enterprise IoT solutions at scale.Download white paper Three Actions Enterprise IT Leaders Can Take to Improve Software Supply Chain Security to learn how and why high-profile software supply chain attacks like SolarWinds and Log4j happened, the key lessons learned from these attacks, as well as actions you can take today to prevent similar attacks from happening to your organization.Running SAP workloads on Google Cloud? Upgrade to our newly released Agent for SAP to gain increased visibility into your infrastructure and application performance. The new agent consolidates several of our existing agents for SAP workloads, which means less time spent on installation and updates, and more time for making data-driven decisions. In addition, there is new optional functionality that powers exciting products like Workload Manager, a way to automatically scan your SAP workloads against best-practices. Learn how to install or upgrade the agent here.Deploy PyTorch models on Vertex AI in a few clicks with prebuilt PyTorch serving containers – which means less code, no need to write Dockerfiles, and faster time to production.Confidential GKE Nodes on Compute-Optimized C2D VMs are now GA. Confidential GKE Nodes help to increase the security of your GKE clusters by leveraging hardware to ensure your data is encrypted in memory, helping to defend against accidental data leakage, malicious administrators and “curious neighbors”. Getting started is easy, as your existing GKE workloads can run confidentially with no code changes required.Week of Feb 3 – Feb 10Immersive Stream for XR leverages Google Cloud GPUs to host, render, and stream high-quality photorealistic experiences to millions of mobile devices around the world, and is now generally available. Read more here.Reliable and consistent data presents an invaluable opportunity for organizations to innovate, make critical business decisions, and create differentiated customer experiences. But poor data quality can lead to inefficient processes and possible financial losses. Today we announce new Dataplex features: automatic data quality (AutoDQ) and data profiling, available in public preview. AutoDQ offers automated rule recommendations, built-in reporting, and serveless execution to construct high-quality data. Data profiling delivers richer insight into the data by identifying its common statistical characteristics. Learn more.Cloud Workstations now supports Customer Managed Encryption Keys (CMEK), which provides user encryption control over Cloud Workstation Persistent Disks. Read moreGoogle Cloud Deploy now supports Cloud Run targets in General Availability. Read moreLearn how to use NetApp Cloud Volumes Service as datastores for Google Cloud VMware Engine for expanding storage capacity. Read moreWeek of Jan 30 – Feb 3Oden Technologies uses BigQuery to provide real-time visibility, efficiency recommendations and resiliency in the face of network disruptions in manufacturing systems. As part of the Built with BigQuery program, this blog post describes the use cases, challenges, solution and solution architecture in great detail.Lytics is a next generation composable CDP that enables companies to deploy a scalable CDP around their existing data warehouse/lakes. As part of the Built with BigQuery program for ISVs, Lytics leverages Analytics Hub to launch secure data sharing and enrichment solution for media and advertisers. This blog post goes over Lytics Conductor on Google Cloud and its architecture in great detail.Now available in public preview, Dataplex business glossary offers users a cloud-native way to maintain and manage business terms and definitions for data governance, establishing consistent business language, improving trust in data, and enabling self-serve use of data. Learn more here.Security Command Center (SCC), Google Cloud’s native security and risk management solution, is now available via self-service to protect individual projects from cyber attacks. It’s never been easier to secure your Google Cloud resources with SCC. Read our blog to learn more. To get started today, go to Security Command Center in the Google Cloud console for your projects.Global External HTTP(S) Load Balancer andCloud CDN now support advanced traffic management using flexible pattern matching in public preview. This allows you to use wildcards anywhere in your path matcher. You can use this to customize origin routing for different types of traffic, request and response behaviors, and caching policies. In addition, you can now use results from your pattern matching to rewrite the path that is sent to the origin.Run large pods on GKE Autopilot with the Balanced compute class. When you need computing resources on the larger end of the spectrum, we’re excited that the Balanced compute class, which supports Pod resource sizes up to 222vCPU and 851GiB, is now GA!Manage table and column-level access permissions using attribute-based policies in Dataplex. Dataplex attribute store provides a unified place where you can create and organize a Data Class hierarchy to classify your distributed data and assign behaviors such as Table-ACLs and Column-ACLs to the classified data classes. Dataplex will propagate IAM-Roles to tables, across multiple Google Cloud projects, according to the attribute(s) assigned to them and a single, merged policy tag to columns according to the attribute(s) attached to them. Read more.Week of Jan 23 – Jan 27Starting with Anthos version 1.14, Google supports each Anthos minor version for 12 months after the initial release of the minor version, or until the release of the third subsequent minor version, whichever is longer. We plan to have Anthos minor release three times a year around the months of April, August, and December in 2023, with a monthly patch release (for example, z in version x.y.z) for supported minor versions. For more information, read here.Anthos Policy Controller enables the enforcement of fully programmable policies for your clusters across the environments. We are thrilled to announce the launch of our new built-in Policy Controller Dashboard, a powerful tool that makes it easy to manage and monitor the policy guardrails applied to your Fleet of clusters. New policy bundles are available to help audit your cluster resources against kubernetes standards, industry standards, or Google recommended best practices. The easiest way to get started with Anthos Policy Controller is to just install Policy controller and try applying a policy bundle to audit your fleet of clusters against a standard such as CIS benchmark.Dataproc is an important service in any data lake modernization effort. Many customers begin their journey to the cloud by migrating their Hadoop workloads to Dataproc and continue to modernize their solutions by incorporating the full suite of Google Cloud’s data offerings. Check out this guide that demonstrates how you can optimize Dataproc job stability, performance, and cost-effectiveness.Eventarc adds support for 85+ new direct events from the following Google services in Preview: API Gateway, Apigee Registry, BeyondCorp, Certificate Manager, Cloud Data Fusion, Cloud Functions, Cloud Memorystore for Memcached, Database Migration, Datastream, Eventarc, Workflows. This brings the total pre-integrated events offered in Eventarc to over 4000 events from 140+ Google services and third-party SaaS vendors.mFit 1.14.0 release adds support for JBoss and Apache workloads by including fit analysis and framework analytics for these workload types in the assessment report. See therelease notes for important bug fixes and enhancements.Google Cloud Deploy Google Cloud Deploy now supports Skaffold version 2.0. Release notesCloud Workstations – Labels can now be applied to Cloud Workstations resources. Release notesCloud Build – Cloud Build repositories (2nd gen) lets you easily create and manage repository connections, not only through Cloud Console but also through gcloud and the Cloud Build API. Release notesWeek of Jan 16 – Jan 20Cloud CDN now supports private origin authentication for Amazon Simple Storage Service (Amazon S3) buckets and compatible object stores in Preview. This capability improves security by allowing only trusted connections to access the content on your private origins and preventing users from directly accessing it.Week of Jan 9 – Jan 13Revionics partnered with Google Cloud to build a data-driven pricing platform for speed, scale and automation with BigQuery, Looker and more. As part of the Built with BigQuery program, this blog post describes the use cases, problems solved, solution architecture and key outcomes of hosting Revionics product, Platform Built for Change on Google Cloud.Pub/Sub Lite now offers export subscriptions to Pub/Sub. This new subscription type writes Lite messages directly to Pub/Sub – no code development or Dataflow jobs needed. Great for connecting disparate data pipelines and migration from Lite to Pub/Sub. See here for documentation.GPU Pods on GKE Autopilot are now generally available. Customers can now run ML training, inference, video encoding and all other workloads that need a GPU, with the convenience of GKE Autopilot’s fully-managed Kubernetes environment.Kubernetes v1.26 is now generally available on GKE. GKE customers can now take advantage of the many new features in this exciting release. This release continues Google Cloud’s goal of making Kubernetes releases available to Google customers within 30 days of the Kubernetes OSS release.Comprehensive guide for designing reliable infrastructure for your workloads in Google Cloud. The guide combines industry-leading reliability best practices with the knowledge and deep expertise of reliability engineers across Google. Understand the platform-level reliability capabilities of Google Cloud, the building blocks of reliability in Google Cloud and how these building blocks affect the availability of your cloud resources. Review guidelines for assessing the reliability requirements of your cloud workloads. Compare architectural options for deploying distributed and redundant resources across Google Cloud locations, and learn how to manage traffic and load for distributed deployments. Read the full blog here.
Quelle: Google Cloud Platform

The year in Google Cloud: Top news of 2023

In the world of technology and cloud computing, the news comes fast and furious. Blink and you’ll miss it. As we wind down the final days of 2023, here’s a look at the top stories of the year that we published on the Google Cloud blog — the product launches, research findings, and initiatives that resonated most with you.JanuaryThe Google Cloud community started the year in a contemplative mood, thirsty for tools to give them deeper insights from their data, and more holistic views of their environments. The top stories of the month were:Log Analytics in Cloud Logging is now GAManage Kubernetes configuration at scale using the new GitOps observability dashboardBetter together: Looker connector for Looker Studio now generally availableIntroducing Security Command Center’s project-level, pay-as-you-go optionsCISO Survival Guide: Vital questions to help guide transformation successFebruaryIn February, readers looked ahead. We unveiled a new pricing approach that is decidedly forward-looking, made inroads on futuristic immersive technology, and planted seeds with the telecommunications community at Mobile World Congress. Readers were also really excited for the global roll-out of AlloyDB. The top stories for the month were:Introducing new cloud services and pricing for ultimate flexibilityExtending reality: Immersive Stream for XR is now Generally AvailableReimagining Radio Access Networks with Google CloudIntroducing Telecom Network Automation: Unlock 5G cloud-native automation with Google Cloud, and Introducing Telecom Data Fabric: Unlock the value of your dataAlloyDB for PostgreSQL goes global with sixteen new regionsMarchIf there has been one overarching theme to Google Cloud for 2023, it’s been generative AI, which made its first real showing this month, with the launch of support for the technology in Vertex AI, alongside an avalanche of news from our first-ever Data Cloud & AI Summit.Google Cloud brings generative AI to developers, businesses, and governmentsNew BigQuery editions: flexibility and predictability for your data cloudBuild new generative AI powered search & conversational experiences with Gen App BuilderIntroducing Looker Modeler: a single source of truth for BI metricsRun AlloyDB anywhere – in your data center, your laptop, or in any cloudAprilWe kicked the AI story up a notch in April with segment-specific news, and a glimpse into our AI-optimized infrastructure. We also made it easier for customers to interact with Google Cloud professional services, and introduced gamified training!A responsible path to generative AI in healthcareSupercharging security with generative AIBringing our world-class expertise together under Google Cloud ConsultingGoogle’s Cloud TPU v4 provides exaFLOPS-scale ML with industry-leading efficiencyBoost your cloud skills — play The Arcade with Google Cloud to earn points and prizesMayGoogle I/O is usually a consumer-focused show, but Google Cloud’s foundational role in enabling generative AI let our news take center stage, including the launch of the Duet AI brand. With that as the backdrop, it’s no surprise readers were also excited about new multicloud connectivity capabilities.Introducing Duet AI for Google Cloud – an AI-powered collaboratorGoogle Cloud advances generative AI at I/O: new foundation models, embeddings, and tuning tools in Vertex AIAt Google I/O, generative AI gets to workAnnouncing A3 supercomputers with NVIDIA H100 GPUs, purpose-built for AIAnnouncing Cross-Cloud Interconnect: seamless connectivity to all your cloudsJuneThree short months after announcing support for generative AI for Vertex AI, we made good by bringing it to general availability, and expanding it to search experiences. We also helped thread the needle between generative AI and databases with vector support, and shared how generative AI is helping to evolve the threat landscape.Generative AI support on Vertex AI is now generally availableHelping businesses with generative AIImproving search experiences with Enterprise Search on Gen App BuilderBuilding AI-powered apps on Google Cloud databases using pgvector, LLMs and LangChain and Announcing vector support in PostgreSQL services to power AI-enabled applicationsExpanding our Security AI ecosystem at Security Summit 2023JulyJuly is usually a relatively quiet month, as people head out on vacations, but not this year — the launch of even more models for our AI builder tools proved just as enticing as a day at the beach. We also shook up the MySQL community with a bold new Cloud SQL Enterprise Plus offering, and introduced a new, visual way for developers to connect their applications.Google Cloud expands availability of enterprise-ready generative AIIntroducing Application Integration: Connect your applications visually, without codeConversational AI on Gen App Builder unlocks generative AI-powered chatbots and virtual agentsIntroducing Cloud SQL Enterprise Plus: New edition delivers up to 3x MySQL performanceAugustIf you think that generative AI news dominated Google Cloud Next this month, you’re only half right. It was certainly a thread in all our leading announcements, but there was also a lot of excitement for more traditional Google Cloud specialties around data analytics and Kubernetes.Announcing BigQuery Studio — a collaborative analytics workspace to accelerate data-to-AI workflowsVertex AI extends enterprise-ready generative AI development with new models, toolingExpanding Duet AI, an AI-powered collaborator, across Google CloudExpanding our AI-optimized infrastructure portfolio: Introducing Cloud TPU v5e and announcing A3 GAIntroducing the next evolution of container platformsSeptemberEnough with the generative AI news already In September, readers remembered that it’s a big world out there, and that Google Cloud is a cloud provider with global coverage. From dashboards to databases, from subsea cables to blockchain, this month’s most popular stories showcased the breadth and depth of Google Cloud’s offerings.Introducing Infrastructure Manager: Provision Google Cloud resources with TerraformMeet Nuvem, a cable to connect Portugal, Bermuda, and the U.S.Enhancing Google Cloud’s blockchain data offering with 11 new chains in BigQueryBigQuery’s user-friendly SQL: Elevating analytics, data quality, and securityGoogle is a Leader in the 2023 Gartner® Magic Quadrant™ for Container ManagementOctoberAround here, we like to talk about having a healthy disregard for the impossible. Like mitigating the largest-ever DDoS attack — again. Or rethinking Ethernet. Or halving the cost of Spanner compared to the competition. Whatevs, no big deal.Google mitigated the largest DDoS attack to date, peaking above 398 million rpsGoogle opens Falcon, a reliable low-latency hardware transport, to the ecosystemShared fate: Protecting customers with generative AI indemnificationCloud Spanner is now half the cost of Amazon DynamoDB, and with strong consistency and single-digit ms latency2023 State of DevOps Report: Culture is everythingNovemberSecurity researchers never sleep, and neither do our systems engineers, uncovering significant new vulnerabilities, and beating records for the world’s largest distributed training job for large language models. Oh, and Memorystore for Redis got an upgrade that delivered 60X more throughput.Google researchers discover ‘Reptar,’ a new CPU vulnerabilityGoogle Cloud demonstrates the world’s largest distributed training job for large language models across 50000+ TPU v5e chipsVertex AI Search adds new generative AI capabilities and enterprise-ready featuresMemorystore for Redis Cluster is GA and provides up to 60 times more throughput and microseconds latencyGKE Enterprise, the next evolution of container platforms, is now generally availableDecemberHalfway through the month, we’re pretty sure we know what the top stories will have been: anything related to Gemini, Google’s latest and most capable AI model. Also a special shout out to Google Cloud’s Learning Content team, whose post about free generative-AI trainings shot up to be the top-viewed post of the entire year — in the matter of a few days. Seems like you are all as excited as we are about the Gemini era!12 days of no-cost training to learn generative AI this DecemberImagen 2 on Vertex AI is now generally availableGemini, Google’s most capable model, is now available on Vertex AIMedLM: generative AI fine-tuned for the healthcare industryAnnouncing General Availability of Duet AI for Developers and Duet AI in Security OperationsAnd that’s a wrap! On behalf of the Google Cloud blog team, wishing you peaceful and happy holiday season, and looking forward to seeing you here on these pages in 2024.
Quelle: Google Cloud Platform

The overwhelmed person’s guide to Google Cloud: week of Oct 23

Quelle: Google Cloud Platform

Getting to know Systems insights, a simplified database system monitoring tool

Quelle: Google Cloud Platform