Changes between Version 5 and Version 6 of Ticket #510, comment 22


Ignore:
Timestamp:
2013-05-30T00:27:12Z (12 years ago)
Author:
daira
Comment:

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #510, comment 22

    v5 v6  
    44 * a sequential scan over blocks will reference the same chunk for several (typically about 12 for k = 3) consecutive requests.
    55 * a single block may span chunks.
     6 * writes not aligned with a chunk must be implemented using read-modify-write.
    67
    78The 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.