The best-case latency on bringing up a new worker on Kubernetes is less than a minute. Kubernetes platform provides us with the capability to add and remove workers from a Presto cluster very quickly. These events enable us to capture the effect of cluster crashes over time.Įach Presto cluster at Pinterest has workers on a mix of dedicated AWS EC2 instances and Kubernetes pods.
When a Presto cluster crashes, we will have query submitted events without corresponding query finished events. Each query is logged when it is submitted and when it finishes. Singer is a logging agent built at Pinterest and we talked about it in a previous post. Within Pinterest, we have close to more than 1,000 monthly active users (out of total 1,600+ Pinterest employees) using Presto, who run about 400K queries on these clusters per month.Įach query submitted to Presto cluster is logged to a Kafka topic via Singer. Presto clusters together have over 100 TBs of memory and 14K vcpu cores. Our Presto clusters are comprised of a fleet of 450 r4.8xl EC2 instances. We have hundreds of petabytes of data and tens of thousands of Apache Hive tables. This separates compute and storage layers, and allows multiple compute clusters to share the S3 data. Our infrastructure is built on top of Amazon EC2 and we leverage Amazon S3 for storing our data. Operating Presto at Pinterest’s scale has involved resolving quite a few challenges like, supporting deeply nested and huge thrift schemas, slow/ bad worker detection and remediation, auto-scaling cluster, graceful cluster shutdown and impersonation support for ldap authenticator. To provide employees with the critical need of interactive querying, we’ve worked with Presto, an open-source distributed SQL query engine, over the years.
0 kommentar(er)
