Welcome

This is a presentation about NoSQL for cartographers.

Why me?

2008: first paid software job.
2012: graduated Computing Science.
2014-2016: Amazon Web Services.
2016+: Uber.
Using SQL and NoSQL at dayjob.

What is a server?

SQL

Invented > 50 years ago.
Technologies differ, concepts are the same.

Power and limitation of SQL

Let's model a house on a piece of land.

There will be always a piece of land under a house.
SQL says you can enforce this.

They must be on the same server to enforce this constraint.

If constraint cannot be created, it's not SQL. If it's SQL, all houses and all areas must be on the same server.

Get a big database

$30 to about a $1M. Show graphics.

Alternatives

Cat videos do not require topology. But we still need storage -- No SQL!

Like-SQL

Forego some SQL constraints, but use the same (/similar) language:

No uniqueness constraints.
No foreign keys.
... many more restrictions.

Hive:

SELECT a.foo FROM invites a WHERE a.ds='2008-08-15';

Key-Value: fast & small

DynamoDB, Cassandra. Some have their language, some don't.

DynamoDB:

table = dynamodb.create_table(
    TableName='users',
    KeySchema=[
        { 'AttributeName': 'username', 'KeyType': 'HASH' },
        { 'AttributeName': 'last_name', 'KeyType': 'RANGE' }
    ],
)

Cassandra:

INSERT INTO monkeys (type, family, common_name, conservation_status)
VALUES ('New World Monkey', 'Cebidae', 'white-headed capuchin', 'Least concern');

Key-Value: slow & big

S3.

https://www.youtube.com/watch?v=8vQmTZTq7nw

1.7 KiB Raw Blame History