Basic implementation of distributed search following the conventions of distributed compression using Celery.

[kirkrodrigues]

Description

Added basic support for distributed search using the celery workers. The algorithm is essentially as follows:

• The search script starts up the search controller which is a TCP server.
• The search script also creates a task for every archive in the database and submits them to the scheduler through the database.
• The scheduler picks up the tasks from the database and queues them for workers.
• A worker picks up a task from the queue, connects to the search controller, runs the search on the corresponding archive and sends any results to the search controller.
• The search controller prints any results it receives.

Validation performed

• Compressed some logs and verified decompression.
• Searched the same logs for some text and verified it returned relevant results.
• Verified that tasks were created for every archive and workers searched the archives in parallel.

[davidlion]

No issues.