Fixed Client bottleneck caused by shared channel and gRPC stub.
Added a synchronous mode client with cache.

Fixed data block size bug for the client

PS. Observing an extra weird null response intermittently, which is currently being handled by a hack

Added some HPDOS architecture diagram

# Conflicts:
#	code/hpdos_client/app.config
#	code/hpdos_client/app/src/main/java/HpdosClient/ClientRunner.java

Added better metrics to client.
Client now configurable through config file

Needs more testing

Preliminary tests down on key create and read

Added Storage model
Added IO handler wrapper class

Added backend for in memory data storage

-- Added a basic interface at MDS to receive gRPC calls from client
-- Added basic infrastructure for RPC implementation
-- Added a client module for E2E implementation testing

Renamed p4 based test codes to tryouts folder

-- Controller IP made dynamic (now set via Control Plane)
-- Also fixed minor controller.py crash bug

-- Enabled caching at switch using netcache
-- Integrated cache with Key partition router
-- Tested partition routing and caching with 2 replicaset
-- Tested functionality of KV working even if server is down

-- Implemented a MDS for routing request to replica set
-- Uses CRC32 as routing algorithm
-- Uses Key in UDP packet payload as the routing criteria