A challenge often faced in multi-tenant SaaS is ensuring that each tenant gets a fair share of a platform's resources. At Salsify, we had to address this in our Delayed Job-based background task execution infrastructure. Because our customers have different use cases, they tend to run tasks of varying complexity and size. Over time we developed tenant-fairness job reservation strategies that made scaling our job system difficult if not impossible. In this post, I discuss how we managed to extend Delayed Job to solve for tenant fairness in a scalable manner.
But despite all of the advantages of running background jobs, under real world usage you can still run into challenging situations that require thoughtful handling. One general class of problems that can arise is achieving fairness in resource usage across users.
Many web interfaces let a user effortlessly page through large sets of data. Implementing database queries that fetch these pages is also effortless for the programmer, usually requiring an
LIMIT in the case of SQL and a
SIZE in the case of Elasticsearch. Although this method is easy on the user and programmer, pagination queries of this type have a high hidden cost for SQL, Elasticsearch and other database engines.
At Salsify, we encountered this problem when implementing a feature to allow a user to step through records in a large, heavily filtered and sorted set. We had to implement an efficient pagination solution that would work in both our SQL and ES datastores. In this post we’ll look at an interesting technique to make pagination efficient for the database with large datasets. Specifically, we’ll look at implementation in SQL as well as how to translate this method to Elasticsearch.