OfferUp's meshing journey from monolith to microservices with Linkerd

Video

May 6, 2019

OfferUp’s meshing journey from monolith to microservices with Linkerd

In this talk, Leo Liang, engineering manager at Cruise Automation, spoke about how OfferUp (his previous employer) evolved a high growth startup architecture into the microservice world through practical architecture examples. Over the past 2.5 years, Leo and his team worked with the Linkered community, leveraged with Consul, Nginx, Prometheus, and deeply customized Linkerd with Plugins to build up the service mesh. While at OfferUp his team scaled to billions of requests per day with ever improving system reliability and flexibility.

REST to gRPC with Linkerd

In this talk, Kevin Lingerfelt, a software engineer at Buoyant and a core contributor to the Linkerd project, shared how companies are moving their architectures from REST-based APIs to gRPC. His talk covered reasons for moving and best practices for running gRPC in production. He detailed on how Linkerd enhances gRPC by providing metrics, load balancing, and support for timeout and retries. He also described how the Linkerd project itself employs multiple gRPC features to facilitate robust communication between its control plane and its data plane.

Mar 28, 2019 | Video

Nordstrom, Kairos, and Attest share application monitoring best practices with Linkerd

In this live discussion moderated by DigitalOcean’s Michelle Peot, we’ll hear from infrastructure platform owners at Nordstrom, Kairos, and Attest on how they use Linkerd as part of their broader application health monitoring strategy to secure, monitor, and observe Kubernetes applications. The conversation covers a lot of ground, including how these organizations have architected their infrastructure, which teams are responsible for application health, how they use Linkerd as part of their application health strategy, and the most challenging aspects of monitoring application health.

Mar 20, 2019 | Video

Strava uses Linkerd to avoid service outages

Strava has been running Linkerd in production for over a year, and almost every internal app uses it for service discovery and routing. From incident response workflows to candidate deployments, Linkerd plays an integral part throughout Strava’s infrastructure stack. In this talk, J Evans, Infrastructure Engineer at Strava will describe Strava’s real-world use case for Linkerd. He will describe how they use Linkerd’s Prometheus and Statsd integrations to respond to service outages, and how these observability tools along with other Linkerd features can even be used to avoid such outages in the first place.

OfferUp's meshing journey from monolith to microservices with Linkerd