WebHEP Analysis Cacheing

Registered by Justin

Cache previous analyses so that users can see them again, and so that previously run configurations don't require computationally expensive analysis.

Blueprint information

Status:
Not started
Approver:
Justin
Priority:
Low
Drafter:
Justin
Direction:
Needs approval
Assignee:
Justin
Definition:
New
Series goal:
Accepted for trunk
Implementation:
Not started
Milestone target:
None

Related branches

Sprints

Whiteboard

Create a unique id for each analysis, probably just a number. Put this in the url on the waiting and plots screen so that a user can bookmark or share it.

Create a DB that has three columns: analysis id, configuration string, results string
  Are long strings appropriate for DBs?
  config strings will have to be made the same even with leading/trailing 0's in input

On run the DB is first queried for the configuration (will this take to long?)
On receipt of a url w/ an ID, DB is queried for id, and plots screen created.
  If id isn't present, return a screen saying analysis not found or deleted from cache.

Will have to delete old/unused entries from DB periodically. Is there a standard tool for this? Maybe also require last look up data column in DB?

Would MongoDB be a good fit? Looks easier to integrate/learn than SQL...
http://www.mongodb.org/display/DOCS/Comparing+Mongo+DB+and+Couch+DB

(?)

Work Items

This blueprint contains Public information 
Everyone can see this information.

Subscribers

No subscribers.