SANTA CLARA, Calif., June 7, 2023 — DataStax, a real-time AI company, today announced that Astra DB, a popular database-as-a-service (DBaaS) built on the open-source Apache Cassandra database, now supports Vector. announced. This is a key feature that enables databases to provide long-term memory for AI applications using Large Language Models (LLM) and other AI use cases.
Following the introduction of vector search in Cassandra, the availability of this new tool in the pay-as-you-go Astra DB service allows developers to build LLMs, AI assistants, and real-time generative AI projects. Goldman Sachs Research predicts that the generative AI software market could grow to he $150 billion, compared to $685 billion in the global software industry.
DataStax will work closely with the Google Cloud AI/ML Center of Excellence as part of the Built with Google AI program to take full advantage of Google Cloud’s generative AI services to enhance capabilities and experience for customers using DataStax doing.
Vector search allows developers to search databases by context and meaning rather than by keyword or literal value. This is done using “embedding” such as Google Cloud’s API for Text Embedding. It can express semantic concepts as vectors for searching unstructured datasets such as text and images.
Embedding is a powerful tool that allows natural language searches across large corpora of data in various formats to extract the most relevant parts of the data.
A vector store is required to enable extremely low latency searches across the database. Combining embeddings, vector stores, and generative AI models like Google PaLM 2, you can create powerful capabilities that dynamically combine the right information in the expected language for the right customer. And now, Cassandra’s semantic search capabilities will play a key role in building generative AI applications.
“Vector search is a critical part of any new AI stack. Every developer building for AI needs to make it easy for AI agents to query their data,” said Ed Anuff, CPO at DataStax. I’m here. “Unlike many other vector databases, Astra DB is not only built with global scale and availability in mind, but it is also the most stringent enterprise database for managing sensitive data, including HIPAA, PCI, and PII regulations. Level requirements are also supported, making this an ideal choice for both startups and enterprises looking to manage sensitive user information and build impactful generative AI applications. will be.”
DataStax is releasing a new vector search tool and other new features via NoSQL Copilot, a Google Cloud Gen AI-powered chatbot that helps DataStax customers develop AI applications on Astra DB. DataStax and Google Cloud release CassIO, an open source plugin to LangChain. This makes it easy to combine his Vertex AI service on Google Cloud with Cassandra for caching, vector search and chat history retrieval.
Stephen Orban, Vice President of Migration and GenAI Ecosystem at Google Cloud, said: “Our customers are always looking for ways to tightly integrate data and AI capabilities.” We will add capabilities and provide our customers with a complete and integrated data and AI solution approach.”
Priceline CTO Martin Brodbeck said: “Vector Search will allow us to semantically query the billions of real-time signals received as part of the checkout experience and returned to Astra DB. We plan to use it in conjunction with vector search to power real-time data infrastructure and generative AI experiences.”
new features
Additionally, DataStax has partnered with Google Cloud on several key new features to further enhance Apache Cassandra and Astra DB as the databases of choice for AI applications.
- Casio – The CassIO open source library makes it easy to add Cassandra to popular generative AI SDKs such as LangChain. Thanks to our close collaboration with Google Cloud, the new integration includes advanced AI assistants, semantic caching for generative AI, LLM chat history, Cassandra prompt templates, and a few key features that enable you to build new Google Cloud Gen AI integrations. features are included.
- Google Cloud BigQuery integration – New integration allows Google Cloud users to seamlessly import and export data from Cassandra to BigQuery directly from the Google Cloud Console to create and serve ML capabilities in real time.
- Google Cloud DataFlow integration – A new integration pipes real-time data to and from Cassandra, providing real-time capabilities to ML models, integrating with other analytics systems such as BigQuery, and monitoring the performance of generative AI models in real time.
Additionally, DataStax has entered into a new IT services partnership to accelerate the use of Cassandra in building real-time generative AI applications.
- SpringML provides advanced data science and AI services. Our close partnership with Google Cloud and early access to Google Cloud’s latest generative AI services ensure we deliver high-quality AI applications to the world’s largest enterprises.
Vector Search is now available in public preview for non-production use on the serverless Astra DB cloud database. Initially available only on Google Cloud, it will be available on other public clouds later. Developers can get started quickly by signing up for Astra.
To learn more about how Cassandra is evolving into an even more powerful database for AI and ML, see our DataStax blog post.
About DataStax
DataStax is a real-time AI company. DataStax enables any enterprise to leverage real-time data to rapidly build smart, high-growth applications at unlimited scale on any cloud. DataStax offers Astra DB cloud database built on Apache Cassandra and Astra Streaming event streaming technology built on Apache Pulsar. Hundreds of the world’s leading companies, including Verizon, Audi, ESL Gaming and more, rely on DataStax to unlock the power of real-time data to capture new markets and transform industries.
Source: DataStax
