Hi remote execution community,
I've finally written up my thoughts on better handling of large blobs within the v2 API.
This schema seems effective for very large blobs, and I believe could be made performant for small blobs though a variety of means.
I'm happy to have your comments and to discuss either on this thread or in the meeting next week.
Thanks,
Steven