Release Notes
Last updated
DA-3.5 (stable)
Last updated
We are excited to announce the next release of Alluxio Enterprise for Data Analytics, Alluxio DA 3.5. It has many noticeable updates to cache management, API improvements, CVE fixes, and operator upgrades.
Directory-Based Quota Management enables users to set resource limits at the directory level. This feature allows administrators to enforce quotas on specific directories. See more information in the Directory-Based Quota documentation.
The new Directory-Based TTL Cache Evicting feature enables users to apply time-to-live (TTL) policies for cache entries based on specific directories. This functionality allows for more granular control over caching, ensuring that data in less frequently accessed directories can be automatically evicted after a designated time period. This feature helps maintain a fresh and relevant cache, optimizing memory usage while prioritizing the retention of important data. For more details, refer to the Directory-Based TTL Cache Evicting documentation.
The Priority-Based Cache Evicting feature enhances the cache management capabilities by assigning a priority level to specific directories. This feature ensures that the more critical data remains in the cache while the less important data can be evicted when memory is needed. This manual demarcation of cached data helps maintain optimal performance for high-priority workloads. Learn more about this feature in the Priority-Based Cache Evicting documentation.
The UFS Read Rate Limiter feature allows users to configure a maximum bandwidth limit per second for UFS reads performed by a single worker. While brief spikes in bandwidth may occur, the average usage is maintained within the specified limit, ensuring controlled data flow to optimize resource utilization and maintain system stability. This feature is particularly useful for managing workloads by capping data processing rates.
Refer to UFS Bandwidth Limiting for more details.
Alluxio Operator now supports configuring heterogeneous worker drives with different sizes. This will provide more flexibility to deploy Alluxio on heterogeneous environments. See the deployment specification in the as well as the configuration option described by the .
HTTP persistent connection (also called HTTP keep-alive), is the idea of using a single TCP connection to send and receive multiple HTTP requests/responses, as opposed to opening a new connection for every single request/response pair. By supporting HTTP persistent connection in Alluxio S3 API, the IO latency of 4KB S3 ReadObject will be decreased by about 40%.
Support multipart upload (MPU)
This release deprecates the previous log4j.properties, replacing it with log4j2.xml as part of upgrading the log4j integration. Please note this is a disruptive change for deployments that were previously configuring logging through conf/log4j.properties; these configurations will be ignored and need to be migrated over to the new xml format for log4j2 in conf/log4j2.xml to take effect.
Several critical CVEs are resolved by removing or upgrading their corresponding packages, such as:
Log4j: Explicitly excluded any log4j 1.x versions that were transitively picked up through various dependencies
Zookeeper: Removed and explicitly excluded from all dependencies, Hadoop related ones in particular
Hadoop: Updated hadoop related dependencies to 3.3.6
Support