To start servers:

python api.py 5000 python api.py 5001 python api.py 5002 python api.py 5003

To run Consistent Hash(While servers on): python consistent_hash.py causes-of-death.csv > consistent_hash_output.txt

To run HRW Hash(Make sure to restart servers(ctrl-c then rerun) to reset datas): python hrw_hash.py causes-of-death.csv > hrw_hash_output.txt

In the first part of the assignment 2, you will be implementing a Flask-Restful application which has the following two endpoints.

Put a new entry into the datastore via HTTP POST. Unique key is a hash of

hash(Year:Cause Name:State) => xxxx

Request

curl http://localhost:5000/api/v1/entries -X POST -d '{"xxxx":"2016,All Causes,All causes,Alabama,52466,920.40
"}' -H "Content-Type: application/json"

Response

201 Created

Retrieve existing entries via HTTP GET.

Request

curl http://localhost:5000/api/v1/entries

Response

200 OK
.
.

{
    "num_entries" : 3,
    "entries" : 
    [
        { "key": "value" },
        { "key": "value" },
        { "key": "value" }
    ]
}

Last, add the Flask server port as a command-line argument so that you can run mulitple servers on different ports.

Example:

python3 api.py 5000
python3 api.py 5001
python3 api.py 5002
python3 api.py 5003

In the part II, you will be writing CSV file parser to read each column from the input csv file. Then, feed the data into:

1. Consistent hashing client
2. Rendezvous (HRW) hashing client

in where you will be uploading data into the RESTful datastore you built in the previous section. In both clients, you will be sharding via HTTP POST to http://localhost:5000/api/v1/entries data into all four instances:

servers = ['http://localhost:5000','http://localhost:5001','http://localhost:5002','http://localhost:5003']

Once you have sharded all data into the cluster, you can finally print out the uploaded data from: http://localhost:5000/api/v1/entries.

Example:

python3 consistent_hash.py causes-of-death.csv
Uploaded all 10296 entries.
Verifying the data.
GET http://localhost:5000
{
    ....
}

GET http://localhost:5001
{
    ...
}

GET http://localhost:5002
{
    ...
}

GET http://localhost:5003
{
    ...
}

python3 hrw_hash.py causes-of-death.csv
...

You might want to implement CSV parser in a separate file called csv_paser.py so that both clients are shared the implementation.