[HADOOP-16696] ABFS Adding an option to Always use Read Ahead, even for non sequential reads

• Type: Sub-task
• Status: Patch Available
• Priority: Major
• Resolution: Unresolved
• Component/s: fs/azure
• Labels: abfsactive
• Flags: Patch
• Assignee: Saurabh
• Reporter: Saurabh
• Created: 08/Nov/19 04:18
• Updated: 27/Feb/24 11:09

Description

Adding a config fs.azure.always.readahead, which is disabled by default, to allow read ahead in case of non-sequential reads, such as when reading parquet file in spark. 

Issue Links

• links to: GitHub Pull Request #1708