• ​Best Practices​

• ​https://aws.amazon.com/dynamodb/getting-started/​

• ​https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/bp-general-nosql-design.html​

Pricing

Reads

1 Read Request Unit: Maximum of 4KB

Writes

1 Write Request Unit: Maximum of 1KB

Node.js

To store Timestamp:

new Date().toISOString()

Designing Your Tables

• Unlike traditional RDBMs, DynamoDB groups sorts by partitions.

  • Each partition returns sorted result.

• Fields used for partition keys are immutable.

  • You cannot change the values for fields which make up the partition key

  • Bad idea to use name as the partition key, because you won't be able to change the name of an object.

Gotchas

Naming

For all things naming regarding DynamoDB.

• DynamoDB doesn't allow you to rename tables

• DynamoDB has Reserved Words, so you'll need to use ExpressionAttributeNames to work around them.

• Table names should be prefixed when deploying. Suggested $service-$stage-$table, for example build-my-prod-events

• Index names don’t need to be prefixed.

• Index names can be reused across tables.For example:

  1. Table Events, indexname: updated at

  2. Table organizations, indexname: updated at

vs RDBMS

DyanmoDB cannot do large "UPDATE queries", for example you cannot do:

UPDATE events
SET status = "past"
WHERE status = "upcoming" AND end_time < NOW();

You have to query each individual event. In this scenario, potentially look at storing the status as a Global Secondary Index. For example:

const response = await dynamodb.queryPromise({
  TableName: process.env.EVENTS_TABLE,
  IndexName: process.env.EVENTS_STATUS_INDEX,
  KeyConditionExpression: "#s = :status AND #e <= :end_time",
  ExpressionAttributeNames:{
    "#s": "status",
    "#e": "end_time", 
  },
  ExpressionAttributeValues: {
    ":status": "upcoming",
    ":end_time": new Date().toISOString(),
  },
  ProjectionExpression: "id",
});
​
// We can use either a single BatchWriteItem or multiple UpdateItems
const promises = response.Items.map((event) =>
  dynamodb.updatePromise({
    TableName: process.env.EVENTS_TABLE,
    Key: {
      id: event.id,
    },
    UpdateExpression: "SET #s = :status",
    ExpressionAttributeNames:{
      "#s": "status",
    },
    ExpressionAttributeValues: {
      ":status": "upcoming",
    },
  })
);
​
await Promise.all(promises);

Serverless

WARNING: Changing the table name in serverless.yml will drop the table.

DynamoDB Shell

• ​http://localhost:8000/shell​

• Works better in Chrome vs Safari

​Reserved Words​

Work around is to use ExpressionAttributeNames and ExpressionAttributeValues

Example of using the keys:

dynamodb.query({
  TableName: process.env.EVENTS_TABLE,
  IndexName: process.env.EVENTS_STATUS_INDEX,
  KeyConditionExpression: "#s = :status AND #e <= :end_time",
  ExpressionAttributeNames:{
    "#s": "status",
    "#e": "end_time",
  },
  ExpressionAttributeValues: {
    ":status": "upcoming",
    ":end_time": new Date().toISOString(),
  },
  ProjectionExpression: "id",
});

Examples of Reserved Words:

• counter

• name

• source

• url

• uuid

• value

Promises (async/await)

const { promisify } = require('util');
...
dynamodb.putPromise = promisify(dynamodb.put);
dynamodb.updatePromise = promisify(dynamodb.update);
dynamodb.getPromise = promisify(dynamodb.get);
...
result = await dynamodb.putPromise({});
result = await dynamodb.updatePromise({});
result = await dynamodb.getPromise({});
​

​GetItem​

// Does not throw an exception if record does not exist
const result = await dynamodb.getPromise({});
​
if (!result.Item) {
  // exit early
}

GetItem cannot be preformed on a Global Secondary Index. You can't use IndexName.

​PutItem​

Will overwrite values if the key is the same; unless you define a ConditionalExpression

If an item that has the same primary key as the new item already exists in the specified table, the new item completely replaces the existing item. You can perform a conditional put operation (add a new item if one with the specified primary key doesn't exist), or replace an existing item if it has certain attribute values. You can return the item's attribute values in the same operation, using the ReturnValuesparameter.

​ConditionalExpression​

docClient.put(
  {
    TableName: 'Example',
    Item: {
      CounterName: 'Example1',
    },
    ConditionExpression: 'attribute_not_exists(CounterName)'
  },
  function (error, data) {
    if (error) {
      // If conflict, an error will be thrown
    } else {
      
    }
  }
);

When using ConditionExpression, and if there is a conflicting.

Sample error:

{
    "message":"The conditional request failed",
    "code":"ConditionalCheckFailedException",
    "time":"2019-02-03T09:20:20.260Z",
    "statusCode":400,
    "retryable":false
}

Query vs Scan

​https://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_Query.html https://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_Scan.html​

Indexes

Sorting is all done on the same partition.

Have to partition wisely.

​https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/bp-indexes-general.html​https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/SecondaryIndexes.html​

Global Secondary Index

Cannot perform GetItem. Use Query or Scan. https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/bp-indexes-general.html​

Reducing DynamoDB costs

• Look for batch fetch

• Look for batch writes

• Look for scans

• Look for large queries

• Check projections

Alternatives

• ​Google Cloud Firestore​

• ​Google Cloud Bigtable​

• ​Google Cloud Datastore​

• ​Azure Table Storage​