Join our daily and weekly newsletters for the latest updates and exclusive content in the industry’s leading AI coverage. Learn more
Google Cloud Announced a A significant number of new features in itself Google Cloud Next Event, last week, with at least 229 new announcements.
Was buried on the Mount of the Incoming News New AI chips and agent ai Abilities, as well as Database UpdatesGoogle Cloud also made some great movements with a large data warehouse service. Among the new opportunities, it is a large unit management that helps organizations to discover, understand and trust the information assets. Management helps to address key obstacles for an EU adoption by ensuring the quality of data, accessibility and reliability.
The stakes are very large for Google as needed for competitors in the information space of enterprises.
BigQuery has been on the market since 2011 and has increased significantly in terms of both opportunities and user database in recent years. As can be seen, BigQuery is also a great job for Google Cloud. He saw that the first time the work was the first time during Google Cloud. According to Google, BigQuery had five times in both snowflakes and the number of customers of verbs.
« This is the first year we allowed to write a customer status that is delicious for me, » Yasmeen Ahmed said, « Managing Director of Information Analysts in Google CULUD told Venturebeat. « DataBricks and Snowflake, they are just another enterprise information storage platforms in the market. We have five times more customers. »
How does Google develop Enterprise’s adoption
When Google now claims to be a larger user base than its opponents, it does not produce his foot from gas. In recent months and especially in Google Cloud, Hiperscaler announced a large number of new opportunities to develop the adoption of the enterprise.
The main problem for the enterprise has the correct information that responds to AI, business service level agreements (SLA). According to Google, according to Gartner research, organizations that do not use the AI use and non-supporting information with the EI’s finished information experience, more than 60% of the AI projects will see and leave more than 60% of the work slash.
This problem stems from three persistent problems for managing the information company:
- Sleeping Information Silosu
- Rapidly changing requirements
- Unequivocal information cultures that do not share the teams around the information around the information.
Google represents a significant separation of management opportunities directly from traditional approaches rather than a large combined management solution, separate tools or processes.
BigQuery Unified Management: Technical Deep Dive
The basis of Google’s announcement is a large management equipped with a new large universal catalog. Unlike traditional catalogs with only basic table and column information, the universal catalog combines three different metadata:
- Physical / Technical Metadata: Scheme recipes, data types and profiles.
- Business Metadata: Business words, drawings and semantic context.
- Time time metadata: Survey patterns, use statistics and format information for technologies such as Apache Iceberg.
This unified approach allows you to provide a comprehensive understanding of a large data asset. What makes the system is particularly strong, Google integrated into a direct management layer by means of a developed AI model, knowledge engine.
Discover the connections between the knowledge engine, the database, enriching the context of working with metadata and automatically increases the activity by developing the quality of monitoring information.
The main opportunities, natural language understanding, automated metadata generation, AI’s strong connection, packaging-related assets, both structured and unstructured data include semantic searches and automatic directory of unstructured data and automatic catalog of automated anomaly.
Forget the estimates, the company is a bigger issue AI
Google strategy exceeds the AI model competition.
« I think most of the industry is aimed at going on this individual leader and really thinks about Google’s problem, » Ahmad said.
This is a comprehensive approach, all enterprise information that answers the life of all enterprise information, for example, how to answer critical questions: How do you deliver? How do you deliver in the measure? How do you deliver on management and security?
By innovating each layer on each layer and bringing these updates together, the instant metadata generation, generation of generation, generation, generation, generation, generation, generation, generation and quality created by Flywheel.
He said that models are important. With the arrival of thinking models like Ahmed, Gemini 2.0, Google has unlocked a large unlock for information platforms.
« When a year ago, when a year before asking a business question, something a little more complicated, in fact, you should take more than one step, » he said. « Suddenly you can get acquainted with a plan with the thinking model … You don’t have to find a way to set up a plan. He knows how to build plans. »
As a result, he said that now a data engineering agent can build a pipeline with three steps or 10 steps. Integrated with Google AI capabilities changed the possible with enterprise information.
Real world effects: How does business benefit
Levi Strauss & Company It offers a compulsory example of how single data management can change operations. The 172-year-old company uses Google’s information management capabilities, because in the first place is a wholesale business to use a business business. In a session on Google Cloud, Levi’s information and Vinay Narayana, who worked for the AI platform engineering, explained the use of the organization in detail.
« We are eager to strengthen our business analysts to access real-time information, » said Narayana. « Before we start our journey to build a new platform, our business users do not know where the information lives and knows the source of the information, if they own the source of the information, there was no document. »
Levi, a data platform on Google Cloud, a data platform organizes information products by business domain, makes them discover through the analytical center (Google’s data market). Each data product is accompanied by detailed documents, generation data and quality measurements.
The results were impressive: « We are 50x faster than our inheritance information platform, which is low. The significant number of visual images are 100x faster, » said Naraya. « There are more than 700 users using the platform now. »
Another example comes from Verizon, which uses a Verizon information initiative to connect combined data between work units as part of a Verizon information initiative.
« It will be the largest Telko data depot in North America, » Arvind Rajagopalan, Information Engineering, Architecture and Products AVP VerizonDuring the next meeting during a Google Cloud, he said.
The company’s data estate consists of about 50 million requests, 35,000 data pipelines and 3,500 users with more than 40 petabytes.
Ahmed also provided numerous modern examples in the Spotlight session on Google Cloud. Radisson Hotel Group recognizes ads in a large way. The teams experienced a 50% increase in productivity, increased by more than 20% income from EU-pighteed campaigns. Gordon Food Service has been ready for AI and data that has increased 96% of applications facing customer
What is the difference in ‘Great’: Explore a competitive view
There are many vendors including Microsoft with Amazon with information, snow flight, Microsoft, Sinaps and Redshift, including an enterprise information storage space. All of these sellers are developing various forms of AI integration in recent years.
Databricks a A comprehensive information lake platform and has been expansion Own AI capabilitiesThank you for a part of the mosaic $ 1.3 billion. Amazon Redshift added support for a generative AI in 2023, along with Amazon Q help users to set up inquiries and get better answers. In turn, the snow is engaged in evolving and Partnership with Great Language Model (LLM) Providers including anthropic.
When Microsoft’s offers are pressed in special proportions, Ahmadin, Sinapse claimed that there was no data platform for the types of cases he used for customers.
« I think that we are leaping the whole industry, because we have worked on all the pieces, » he said. « We have the best model, by the way, the best model is the best model in a data stack that agents work. »
This integration has managed rapid acceptance of AI capabilities within the bigquery. According to Google, the use of Google’s AI models for multimodal analysis has increased 16 times a year.
What does this mean for the enterprises accepted by AI
For businesses that are already fighting the AI, Google’s integrated approach can provide a separate data management and a more relevant way to bring together AI systems.
Ahmad’s opponents in this place of Google checks these new opportunities in this place as the « leash » opponents put these new opportunities. However, customer samples and technical details offer Google to make a significant progress in resolving one of the most difficult parties of the AI’s reception.
For technical decision makers that evaluate the information platform, this integrated approach gives a significant value to transfer to existing platforms, and Google’s competitors can protect the current innovation rate?
Source link
Leave a Reply