BIND 10 #2875: Implement the multi-process and layered cache approach

[BIND 10 Development]

#2875: Implement the multi-process and layered cache approach
----------------------------------------+----------------------------------
                   Reporter:  vorner    |                 Owner:
                       Type:  task      |                Status:  new
                   Priority:  medium    |             Milestone:  New Tasks
                  Component:  resolver  |              Keywords:
               CVSS Scoring:            |             Sensitive:  0
            Defect Severity:  N/A       |           Sub-Project:  DNS
Feature Depending on Ticket:            |  Estimated Difficulty:  0
        Add Hours to Ticket:  0         |           Total Hours:  0
                  Internal?:  0         |
----------------------------------------+----------------------------------
 On top of #2871 implement the approach with multiple processes. Each
 process works with coroutines. There are several layers of bigger and
 bigger caches (the smaller are shared with less processes, the digger ones
 with more). Upper cache is mapped read-only into memory and protected by
 RCU (so reading is lock-less). If the data is not in the cache, the cache
 is asked to obtain it somehow by some kind of IPC. The cache gets it
 either from its parent cache or from network, updates itself and notifies
 back down.

 Experiment with number of layers and how to share the caches.

 Tip: We don't really need to do the mapping of memory here. We mostly need
 to test how expensive the IPC would be.

 More information can be found in
 https://lists.isc.org/pipermail/bind10-dev/2013-March/004493.html.

-- 
Ticket URL: <http://bind10.isc.org/ticket/2875>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development