Windkit Li

Throughout the project there are many possibilities to exploit parallelism. Interleaving read, send, calculation and etc. it would be great if we can come up with a framework to achieve...

@mocchira do you have any good ideas to list out all eleveldb instances in a node?

To clarify, `boto` uses `List Parts` to find out parts being uploaded to server http://docs.aws.amazon.com/AmazonS3/latest/API/mpUploadListParts.html I closed the duplicate entry and would continue the discussion here while I am working...

We have done the test back to old 1.4.0 series https://github.com/leo-project/notes/tree/53926f8d26b4a0ada9d9917e38af0a1fd5ac3144/leofs/benchmark/leofs/1.4/1.4.0/erlang_options/20151119_1m_r95w5_30min_1 The difference between default options and suggested options is little (if any) I think we considered to drop the...

So to solve this, we need to keep track the fileid -> path, if a collision is found we can do double hashing. As the table could be accessed by...

First of all, let me confirm some terminology here. The FileID as a unique identifier is for client to do caching and so on. Therefore, for each file, it should...

If you are worrying about collision in 64 bit hash value, I guess the memory usage for string integer should also be considered? That's what I meant with And it...

First of all, I want to ask how bad was it. We did solve similar problem with large object write handler in 1.3.1 https://github.com/leo-project/leofs/issues/570 At the same time, I would...

I have confirmed the issue exists when the GW bandwidth is limited to 10MB/s (in 100MB/s it works as before)

While I am still checking the root cause, there is something I don't feel right in `leo_large_object_put_handler` https://github.com/leo-project/leofs/blob/master/apps/leo_gateway/src/leo_large_object_put_handler.erl#L309 when all the `leo_pod` workers are out, processes would wait 30 seconds...