Parallel preparation, the parallel stream will open the file when reading from it and seek to the right offset to read, and close the file after reading.
We checked it has no performance impact with mmap impl, and it's simpler. Simpler is better
With this approach, the user of the parallel input stream will have to manage the threads and concurrence, but it's only used by our s3 client, which already handles the concurrence and thread pool, we are good.
Parallel preparation, the parallel stream will open the file when reading from it and seek to the right offset to read, and close the file after reading.
We checked it has no performance impact with mmap impl, and it's simpler. Simpler is better
With this approach, the user of the parallel input stream will have to manage the threads and concurrence, but it's only used by our s3 client, which already handles the concurrence and thread pool, we are good.
Original: https://github.com/awslabs/aws-c-s3/pull/353
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.