andres_mogollon_q3_lru_cache_test_ormuco

#IT Arquitecture 1 Central server (CS) 2 Distributed servers (DS) 3 Client (Cl)

#How it works

1. Run the Central Server in port 20000

command: python central_server.py -p 20000

this Central Server register all the Distributed servers when they get connected

• this implementation allows just 3 connections from distributed servers for this test. You can change this limit in central_server.py line 18

2. Run 2 Distributed Server (you can run as much as you want) in the port you need

command: python server.py -p 3001 command: python server.py -p 3002

these servers are going to register their ip and port in the Central Server

• the Distributed Server will have a size cache (3 elements) for this test. You can change this limit in server.py line 114

3. Run the client

command: python client.py

the client asks to the Central Server the list of available servers and a word, then with the list of servers calculates the distance and access to the nearest server requesting data according to the word that client is looking for words to test: network, center, born

4. When the distributed server receives the connection, it looks for the word in the cache and if not there, just looks for it in the dictionary util/db.py and then update the cache

I used this repo as a reference

https://github.com/samisajid/Geo-distributed-LRU-cache

Considerations

the test will be executed in localhost but different ports.

whats done

1 - Simplicity. Integration needs to be dead simple. 2 - Resilient to network failures or crashes. 4 - Data consistency across regions 5 - Locality of reference, data should almost always be available from the closest region

whats missing

3 - Near real time replication of data across Geolocation. Writes need to be in real time. 6 - Flexible Schema 7 - Cache can expire

if a server crashes the central server would have to update the server list to get the performance better