DISCO Improves Search Times and Results Using AWS

Initially conceived in 2013 at a litigation boutique in Houston, Texas, DISCO was created after its founders grew frustrated with the available e-discovery software, which was slow and difficult to use. This prompted them to develop their own solution: DISCO Ediscovery. The company now offers complete solutions with tools that automate and simplify the discovery process so that lawyers can focus on the tasks that require legal judgment.

The main challenge that DISCO’s law firm and corporate law department customers face is managing the wide range of data during discovery. “Document collection can contain 1,000 documents, or it could be 50 million documents,” says Jim Snyder, chief architect at DISCO. “We have to deal with a large variety of scale from all different kinds of clients.” An additional challenge was the diversity of file formats. There are over 2,000 different file formats, and the format can vary widely when documents come from software available as far back as the 1980s. DISCO used the scalability and elasticity of cloud-native AWS services to manage these varied workloads and complexities so that it could provide a consistent experience for its users.

Since 2015, DISCO has been using AWS services to develop its solutions with the flexibility of the cloud. In particular, the company employs AWS Lambda to manage the scale of data that its clients need. Using AWS Lambda, DISCO’s solution can quickly ingest and process large volumes of data to render quality results, giving users a fluid and responsive experience. “If I have to wait a few seconds to turn a page in a document on the computer, that’s going to interrupt my train of thought,” says Snyder. “When you’re in a billable-hour industry, a few seconds versus a few milliseconds is a big difference. You want to let humans move at the rate that they work as opposed to waiting for the machine.” In addition, lawyers often gain access to documentation based on data found during discovery, which creates waves of documents that come in at different times. Using AWS Lambda in the data ingestion and delivery process, DISCO can handle the inconsistent workload without sacrificing response time. The company’s goal is for ingestion to remain consistently fast, regardless of the volume of data.

For use cases where AWS Lambda isn’t a good fit, DISCO uses AWS Fargate, a serverless compute engine that lets organizations focus on building applications without managing servers. “This is an example of where the range of services that AWS provides helps us adapt our architecture,” says Snyder. DISCO can choose which AWS services to use based on the specifics of each solution and the changing needs of its clients.

In addition, DISCO offers data enrichment through artificial intelligence. For example, the company uses algorithms to reconstruct email chains from different mailboxes to help lawyers understand the cohesive conversation. DISCO’s artificial intelligence makes the user experience more immersive and makes Elasticsearch smarter. DISCO’s elastic clusters use Amazon Elastic Compute Cloud (Amazon EC2), a web service that provides secure, resizable compute capacity in the cloud. The scale that DISCO uses is high: one elastic cluster can have over 1,000 nodes and can invoke AWS Lambda billions of times per month.

For its database layer, DISCO uses Amazon Aurora, a MySQL- and PostgreSQL-compatible relational database built for the cloud that provides the performance and availability of commercial-grade databases at a fraction of the cost. “The databases that we can get now from AWS really back our ability to search and materialize data, and the management infrastructure that we get from AWS is really hard to replicate in an on-premises situation,” says Snyder. “We rely on Aurora and Amazon EC2 for running our elastic clusters so that once we’re done processing all these documents, the lawyers can get access to this high-quality information rapidly.”

DISCO employs a distributed domain-driven design built on AWS so that the company can adjust and upgrade as necessary, which helps make it agile and future ready. Alongside this architecture, it uses Amazon Managed Streaming for Apache Kafka (Amazon MSK)—a fully managed, highly available service to securely stream data—to provide the messaging performance the company needs without requiring staff to spend time maintaining it. With its distributed domain-driven design, DISCO can quickly adapt to the changing needs of its clients and adopt new services. “Because of the constant pace of innovation at AWS, we can build more features to meet our customers’ ever-evolving needs,” Snyder says.

Using AWS Lambda and other AWS services, DISCO can provide the fast search and response times that lawyers and legal teams rely on to do their work. For the company’s clients, the speed of searches and quality of results offer a high return on investment, especially when a small team needs to process a large volume of data. “We’ve had boutique firms say that they won a certain case and that there’s no way that they could have taken on a case of that size without DISCO,” says Snyder. “AWS is an integral part of what we do and how we think about design.”