HomeServeNow Enhances App Performance with Thundra and AWS

HomeServeNow is a startup that is building a platform to reinvent how homeowners find certified, available, and affordable tradesmen for all home repairs and improvement needs—within one hour. To ensure its applications enabled fast customer connections, HomeServeNow leveraged serverless compute from Amazon Web Services (AWS), a microservices-based architecture, and performance insights from Thundra. As an observability and security solution, Thundra offered HomeServeNow a clear indication of bottlenecks within its applications, speeding the time to reproduce, respond to, and fix bugs.

The backbone of HomeServeNow is an application that allows tradespeople to bid on and obtain home repair jobs. With a microservices-based architecture, HomeServeNow was ready to quickly scale and accommodate more requests from new customers. The company decided to go completely serverless with AWS Lambda, enabling them to develop and deploy its application quickly, without worrying about server allocation or management. Ensuring fast and responsive application and performance at scale was key, allowing tradespeople to find out—in real time—whether they won or lost a bid. However, HomeServeNow was finding it difficult to troubleshoot any reported issue on production quickly, as the tooling around logging and performance insights were limited. To address this, it turned to Thundra, an AWS Advanced Technology Partner. 

Thundra, with its log aggregation, search capabilities, and performance insights, enabled HomeServeNow to pinpoint bugs and inefficient resources that caused costly lags. Thundra, specializing in serverless-centric workloads, helped HomeServeNow speed its mean time to response and consolidated multiple costly, disconnected tools for application monitoring.

Through in-depth performance analyses from Thundra, HomeServeNow reduced invocation durations by 50 percent, mean time to response by 60 percent, and troubleshooting time by 90 percent.

Along with clearly indicating bottlenecks within HomeServeNow applications, Thundra granted valuable insights into how various services were talking to each other. With Thundra analyses, HomeServeNow was able to aggregate logs, observe processes, and troubleshoot any needs within the serverless domain. Furthermore, Thundra’s distributed tracing and debugging capabilities empowered HomeServe- Now’s teams to proactively triage production issues—before they ever reached a customer.

Thanks to Thundra, HomeServeNow was able to diagnose the cause of slow response times within its end services. For instance, when AWS Lambda had timed out, Thundra immediately alerted the team through Slack, enabling them to quickly pinpoint the root cause of a database table that wasn’t properly structured. Instead of stepping through lines of code to resolve the issue, Thundra provided error logs indicating which service interaction was causing problems. From there, HomeServeNow teams could develop the fix quickly, and deploy to production. Without Thundra, HomeServeNow would lose valuable time deciphering why application performance was subpar.

Performance insights via Thundra has better empowered HomeServeNow to prioritize fixes. Thundra gave insights on the number of invocations and compute costs of service interactions, enabling HomeServeNow to prioritize high-value service interactions first.

Via the powerful combination of AWS Serverless Resources (AWS Lambda, AWS API Gateway) and Thundra, HomeServeNow has the agility to scale for new customers, and the developer velocity to deploy new features fast. Teams that once took up to two days to deploy a new service can now deploy within 10 minutes. Developers aren’t burdened by provisioning servers, and can focus on their specific function in one programming language.