Elasticsearch enables next-gen search with Azure OpenAI Service

The need for a managed cloud-based search solution

No one wants to sift through massive amounts of data to find what they need. For enterprises and smaller businesses alike, the ability to search internal systems and locate information quickly is essential to success.

Elastic specializes in search technology that helps customers, employees, and organizations locate information faster while ensuring data security and keeping mission-critical applications running smoothly. Founded in 2012, the company serves customers in a wide range of industries, including manufacturing, retail, and the public sector.

Soon after releasing its self-managed solution, Elastic recognized the need to build a cloud-based version. “We knew our customers would prefer a managed service because it’s easy to use and gives them more time to focus on their customers,” says Steve Kearns, Vice President of Product Management at Elastic. 

The company wanted to create an easy-to-operate, scalable cloud solution that maintains high data security and allows users to manage how data is shared. The solution also needed to process large amounts of information quickly—and make customers’ data searchable across on-premises and cloud-based systems.

Meeting customers in the cloud

In 2019, the company introduced Elastic Cloud, built using Microsoft Azure. The solution allows Elastic to handle many basic management services for its customers, such as ensuring best practices for data backups, handling upgrades, and growing or shrinking deployments. “Our customers would have to do all of these things if they were self-managing the software, but now we can handle them in a cloud offering through Microsoft Azure,” says Kearns. 

Elastic decided to build a managed solution on Azure partly because its customers were already using it. “It made sense to us to have our query workloads running near where the rest of our customer data exists,” Kearns says. “Many of our customers are running on Microsoft Azure, so we wanted to bring the service to them.”

Because Elastic Cloud users don’t have to manage the granular details of software operation, they can focus time and energy on using the tools and engaging with their own customers. When companies are moving workloads from on-premises to cloud, Elastic eases that transition with hybrid cross-cluster search. “Because our products are available to run in their own data center and also through the Azure Marketplace, our customers can use the same software to bring together on-premises and cloud data in a single pane of glass,” says Kearns. “It means they have a much easier time when they’re in digital transformation mode.”

In addition, Elastic Cloud on Kubernetes (ECK) empowers Elastic’s customers to provision, manage, and operate Elasticsearch clusters on Kubernetes. They can choose to run Microsoft Azure by deploying ECK on Azure Kubernetes Service (AKS) to develop and deploy cloud-native apps, reaping the benefits of AKS’s built-in code-to-cloud pipelines and guardrails.

Rapid growth and satisfied customers

Since introducing its cloud-based solution on Azure, Elastic has seen significant success. “Our cloud business is growing at twice the rate of the rest of the business,” says Kearns. “Elastic Cloud on Azure is very effective for us.”

The cloud offering is transformational for Elastic’s customers. Kearns says that one global telecom provider achieved almost 300 percent return on investment using Elastic Cloud. A recent survey of a few hundred customers revealed that, on average, revenue through system availability increased by 82 percent and customer employee satisfaction for enterprise search users jumped by 74 percent. 

As Elastic continues to improve its products and services, it appreciates the opportunity to work closely with Microsoft. “Having a direct connection to the Azure development team lets us understand their road map and work together to maximize effectiveness around compute, storage, and networking across the board,” Kearns says. “It’s been really helpful to ensure we’re making the most of these first-party services that we rely on from Azure.” In 2023, Elastic won two Microsoft Partner of the Year awards for the commercial marketplace, highlighting their innovation based on Microsoft technology and the value they provide to customers.

Exploring new possibilities with generative AI

The emerging world of generative AI promises to unlock a wave of possibilities for Elastic users. Incorporating Microsoft Azure OpenAI Service, Elastic Cloud can offer more relevant answers to search queries with the Elasticsearch Relevance Engine (ESRE). ESRE enables text, vector, and hybrid search—and includes a vector database, semantic search with Elastic’s sparse encoder model, and support for multiple NLP models. Instead of returning a list of links to internal documents, Azure OpenAI Service processes all available information to create a focused response. “That ability to fold in your company’s internal knowledge with context from actual documents makes generative AI far more effective and more accurate,” says Kearns. 

Using ESRE, Elastic Cloud customers in many verticals are experimenting with Azure OpenAI-powered search possibilities and applications. One example is Relativity, a legal search software company that is a Microsoft and Elastic customer. “When legal professionals do a search, they don’t want a general answer—they want to find the relevant documents that are important to a particular case and to summarize what’s different about them in relation to other cases,” says Kearns. “ESRE and OpenAI can deliver those accurate, nuanced responses.” 

“With Elastic’s ESRE, we see the potential of providing a search experience that goes beyond just the keyword searching,” says Brittany Roush, Product Manager at Relativity. “It provides an opportunity to create an experience that’s augmented with numerous AI capabilities such as GPT4, signals, classifications, and our own in-house AI solutions.”

For the Elastic team, the ability to incorporate emerging technology and deliver even better search tools is exciting. “We’ve been building toward this moment for many years, and that investment has allowed us to really step out in front when it comes to the capabilities needed in the space.”

Find out more about Elastic on Twitter, Facebook, and LinkedIn.