Context Navigation

Changes between Version 6 and Version 7 of Ticket #510, comment 22

v6	v7
6	6	* writes not aligned with a chunk must be implemented using read-modify-write.
7	7
8		The cloud backend uses [https://github.com/LeastAuthority/tahoe-lafs/blob/1c1aff2b58c121dbf74a781525d3f65060deb54d/src/allmydata/storage/backends/cloud/cloud_common.py#L575 caching] to mitigate any resulting inefficiency. However, this is only of limited help because the storage client lacks information about the behaviour of the chunk cache, and the storage server lacks information about the access patterns of the uploader or downloader.
	8	The cloud backend uses [https://github.com/LeastAuthority/tahoe-lafs/blob/1c1aff2b58c121dbf74a781525d3f65060deb54d/src/allmydata/storage/backends/cloud/cloud_common.py#L575 caching] to mitigate any resulting inefficiency. However, this is only of limited help because the storage client lacks information about where the chunk boundaries are and the behaviour of the chunk cache, and the storage server lacks information about the access patterns of the uploader or downloader.
9	9
10	10	A possible performance improvement and simplification that I'm quite enthusiastic about for an HTTP-based protocol is to make blocks the same thing as chunks. That is, the segment size would be k times the chunk size, and the uploader or downloader would directly store or request chunks, rather than blocks, from the backend storage, doing any caching itself.