People asked me what are the recent performance improvements in HDFS. I used to maintain a spreadsheet of these improvements, and I just updated it to incorporate the recent one.
If you're interested, feel free to hover to https://docs.google.com/spreadsheets/d/1dvLoZ039ZirdZF9p0wWKhFCtD91jfbdkPg4XZ-AnMNg/edit?usp=sharing It might be useful to publish it as a browsable, searchable place, like the Apache wiki. I'll try to make that happen. I am thinking of writing a blog post to introduce these good stuff made by our awesome community contributors, but of course I don't have all the details so I plan to make a draft for you to review. Best, Weichiu