If you implemented Artifactory HA versions prior to 5.0 with S3 storage, you needed a cluster-wide write cache (called the eventual cache) to be implemented on a shared NFS mount. Jfrog recommend’s EFS in Provisioned Throughput mode for this cache as it typically has low storage and high usage, which aligns with the workload type that EFS Provisioned Throughput mode was designed to support. For Artifactory versions 5.0 and higher, the cache for an S3 implementation should be EBS local disk caches.