Citrix Case Study

Neo4j Keeps Millions of Remote Workers on Citrix Safe, Productive, and Secure

Citrix is essential to remote work, with a goal to enable users across more than 16 million virtual desktops and more than 150 million devices to be productive across any virtual environment. With more growth and mobility came the need for greater security, threat detection, performance monitoring, and scale, a critical role driven by Citrix Analytics.

However, as Citrix’s Analytics Data Platform expanded, limitations emerged. Analyzing relationships between the growing number of highly connected data points stored in tables became difficult. With data complexity increasing across millions of entities, Citrix faced challenges to efficiently store, manage, and draw insights from their connected data.

That’s when Citrix decided to turn to the Graph Database & Analytics leader.

Neo4j Powers Citrix Analytics to Navigate Complex Relationships

Citrix needed to store and analyze the complex relationships between its growing customer base and various applications and services. Citrix chose Neo4j for its mature, proven graph database solution with enterprise-grade customer support to manage these extensive data relationships.

Unlike relational databases, Neo4j’s native graph data model treats the highly connected multi-level relationships between entities as data itself, making its capture and analysis easy, scalable, and performative. The result has enabled more advanced analytics, enabling Citrix to better predict emerging threats vs simply managing them. Citrix appreciated that data without this depth of context isn’t nearly as accurate or useful when it comes to improving threat detection.

“Initially, we tried to store information in relational databases, but it became difficult to manage because those relationships are quite deep, and doing analysis on the basis of relationships between tables became limiting for us,” said George Tsolis, Distinguished Engineer at Citrix and Chief Architect of Citrix Analytics.

Multi-Tenancy Drives Data Integrity and Privacy

Graph databases are particularly valuable for organizing and giving context to a wide range of data entities. Graph for example can be used to model intricate relationships between entities in the Citrix environment, including relationships between users, devices, virtual applications and desktops, and locations across the Citrix ecosystem. This means that Citrix Analytics can deliver data on how users are actually accessing their environment in real time, enabling them to identify any potential security risks and address them right away – as well as scale this capability as its customers scale.

“We needed ways to partition the graph database so that we could execute queries for its different customer tenants but not have to create separate graph databases for each tenant. So multi-tenancy was essential,” Tsolis said. “The ability to store millions of entities and the respective number of relationships was important as well — as was executing queries to get results in sub-second response times, all while scaling.”

Privacy in the form of isolating data between different customer tenants was also critical. The team needed to store everything in one multi-tenant graph while ensuring a reliable way to keep the sensitive data of all customers partitioned. Maintaining the integrity of its data was a key product requirement; Citrix could not risk a customer logging into Analytics and seeing the data of another customer.

“With Neo4j’s high-availability architecture, we are able to deploy core servers to three Azure availability zones and leverage automatic failure recovery. This is one of the main reasons why we chose Neo4j, along with the confidence in the data integrity that Neo4j provides,” said Jingshi Tu, Product Development Manager for Citrix Analytics. If one zone goes down, the other zones help ensure continuous uptime and availability, preventing downtime and data loss.”

Expanding Graph: Citrix’s Vision for Enhanced Security and Beyond

The Citrix graph today has 25M nodes, 150M edges, is queried 40K times an hour – with plans to leverage the power of graph even further. The result has made Neo4j central to the Citrix Analytics application, and Tsolis is now applying graph to other use cases particularly as it relates to security.

“One avenue we’re moving toward is leveraging the graph database, not only as a component of security analytics but as a key service to our data platform that allows other Citrix products to store and retrieve rich security context in that graph,” Tsolis said.

Get in touch

Curious about what insights you could unlock for your business with graph-powered solutions? Let’s talk – reach out, and we’ll get in touch.