Closed Bug 551383 Opened 14 years ago Closed 11 years ago

Process data from multiple collections instead of one collection at a time

Categories

(Firefox :: Sync, enhancement, P5)

enhancement

Tracking

()

RESOLVED WONTFIX
Future

People

(Reporter: Mardak, Unassigned)

References

Details

(Whiteboard: [storage-api-addition?])

This helps mobile a lot because right now we do requests per collection instead of asking for "all new stuff." So this would reduce the number of connections we make per sync.

Bug 551382 adds the path to fetch/write. WBOs will need to be updated to set a collection attribute.
Instead of doing service start, engine1 start/download/process/upload/end, engine2 start/download/process/upload/end, service end..

We would do service/engines start, download in one batch, process 0+ times, upload in one batch, end.

Engines need to split out code that happen once per sync vs one per process-incoming such as caching data for sync vs turning on/off batch mode.

Data downloaded in one batch can be written directly to disk and then we can nsILineInputStream.readLine one at a time and JSON.parse and hand it off to the appropriate engine.
Target Milestone: 2.0 → ---
Severity: normal → enhancement
Priority: -- → P5
Target Milestone: --- → Future
Whiteboard: [storage-api-addition?]
This requires storage server support.

telliott: do we have any intent of supporting this on the server?
Not in 2.0
Real-world experience doesn't suggest that this is important. If we want to solve something similar to this, it'll be a smarter kind of batching.
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → WONTFIX
Component: Firefox Sync: Backend → Sync
Product: Cloud Services → Firefox
You need to log in before you can comment on or make changes to this bug.