This document discusses using a multi-model database like ArangoDB for microservices. It explains how ArangoDB can store different data models like key-value, documents, and graphs to support microservices that use different data structures. It provides examples of breaking up a monolithic application into microservices that use different parts of the database, and using Foxx to build REST APIs on top of ArangoDB to integrate microservices.

1. www.arangodb.com
NoSQL meets Microservices
Michael Hackstein
@mchacki

2. Michael Hackstein
‣ ArangoDB Core Team
‣ Web Frontend
‣ Graph visualisation
‣ Graph features
‣ Host of cologne.js
‣ Master’s Degree
(spec. Databases and
Information Systems)
2

3. Classical Setup
3

4. Scaling
4
LoadBalancer

5. Responsibilities
5
LoadBalancer
Application Code
Consistency
Joins

6. Introduction to NoSQL
6
NoSQL World
Documents - JSON
Graphs
Key Value
{
“type“: "pants",
“waist": 32,
“length”: 34,
“color": "blue",
“material”: “cotton"
}
{
“type“: "television",
“diagonal screen size": 46,
“hdmi inputs": 3,
“wall mountable": true,
“built-in digital tuner": true,
“dynamic contrast ratio”: “50,000:1”,
Resolution”: “1920x1080”
}
{
“type": "sweater",
“color": "blue",
“size": “M”,
“material”: “wool”,
“form”: “turtleneck"
}
{
“type": "sweater",
“color": "blue",
“size": “M”,
“material”: “wool”,
“form”: “turtleneck"
}
‣ Map value data to unique string keys (identifiers)
‣ Treat data as opaque (data has no schema)
‣ Can implement scaling and partitioning easily
‣ Focussed on m-to-n relations between entities
‣ Stores property graphs: entities and edges can have
attributes
‣ Easily query paths of variable length
‣ Normally based on key-value stores (each document still
has a unique key)
‣ Allow to save documents with logical similarity in
“collections”
‣ Treat data records as attribute-structured documents
(data is no more opaque)
‣ Often allow querying and indexing document attributes

7. Polyglot Modeling
‣ If you have structured data
➡ Use a document store
‣ If you have relations between entities and want to efficiently query
them by arbitrary steps in between
➡ Use a graph database
‣ If you manage the data structure yourself and do not need
complex queries
➡ Use a key-value store
‣ If you have structured data in graph format, or data model might
change
➡ Use a multi-model database
7
Use the right tool for the job

8. Using different databases
8
LoadBalancer

9. Responsibilities
9
LoadBalancer
Application Code
Consistency
Joins

10. Responsibilities
9
LoadBalancer
Application Code
Consistency
Joins

11. Solution 1: Writes only to one master
10

12. Solution 2: Microservices
11

13. app
Monolith breakdown
12
server.route({
method: "GET", path: "/app",
handler: function (req, reply) {
var sum = cpuWaste();
reply(dataLookup["key" + sum]);
}
});
var cpuWaste = function() {
var sum = Math.floor(Math.random() * 10000);
for (var j = 0; j < 1000000; ++j) {
sum += j; sum %= 10000;
}
return sum;
};
var dataLookup = {};
for (var j = 0; j < 500000; ++j) {
dataLookup["key" + j] =
"This is test text number " + j + ".";
}

14. memService
cpuService
app
Monolith breakdown
12
server.route({
method: "GET", path: "/app",
handler: function (req, reply) {
var sum = cpuWaste();
reply(dataLookup["key" + sum]);
}
});
var cpuWaste = function() {
var sum = Math.floor(Math.random() * 10000);
for (var j = 0; j < 1000000; ++j) {
sum += j; sum %= 10000;
}
return sum;
};
var dataLookup = {};
for (var j = 0; j < 500000; ++j) {
dataLookup["key" + j] =
"This is test text number " + j + ".";
}

15. memService
cpuService
app
Monolith breakdown
12
server.route({
method: "GET", path: "/app",
handler: function (req, reply) {
var sum = cpuWaste();
reply(dataLookup["key" + sum]);
}
});
var cpuWaste = function() {
var sum = Math.floor(Math.random() * 10000);
for (var j = 0; j < 1000000; ++j) {
sum += j; sum %= 10000;
}
return sum;
};
var dataLookup = {};
for (var j = 0; j < 500000; ++j) {
dataLookup["key" + j] =
"This is test text number " + j + ".";
}
server.route({
method: "GET", path: "/mem/{id}",
handler: function (request, reply) {
reply(dataLookup["key" + request.params.id]);
}
});

16. memService
cpuService
app
Monolith breakdown
12
server.route({
method: "GET", path: "/app",
handler: function (req, reply) {
var sum = cpuWaste();
reply(dataLookup["key" + sum]);
}
});
var cpuWaste = function() {
var sum = Math.floor(Math.random() * 10000);
for (var j = 0; j < 1000000; ++j) {
sum += j; sum %= 10000;
}
return sum;
};
var dataLookup = {};
for (var j = 0; j < 500000; ++j) {
dataLookup["key" + j] =
"This is test text number " + j + ".";
}
server.route({
method: "GET", path: "/mem/{id}",
handler: function (request, reply) {
reply(dataLookup["key" + request.params.id]);
}
});
server.route({
method: "GET", path: "/cpu",
handler: function (request, reply) {
reply(cpuWaste());
}
});

17. memService
cpuService
app
Monolith breakdown
12
var cpuWaste = function() {
var sum = Math.floor(Math.random() * 10000);
for (var j = 0; j < 1000000; ++j) {
sum += j; sum %= 10000;
}
return sum;
};
var dataLookup = {};
for (var j = 0; j < 500000; ++j) {
dataLookup["key" + j] =
"This is test text number " + j + ".";
}
server.route({
method: "GET", path: "/mem/{id}",
handler: function (request, reply) {
reply(dataLookup["key" + request.params.id]);
}
});
server.route({
method: "GET", path: "/cpu",
handler: function (request, reply) {
reply(cpuWaste());
}
});
server.route({
method: "GET",
path: "/app",
handler: function (req, reply) {
request('http://127.0.0.1:9000/cpu',
function (error, response, body) {
if (!error
&& response.statusCode == 200) {
request('http://127.0.0.1:9000/mem/'
+ body,
function (error, response, body) {
if (!error
&& response.statusCode == 200) {
reply(body);
} else {
reply({error: error});
}
});
} else {
reply({error: error});
}
});
}
});

18. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

19. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

20. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

21. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

22. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

23. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

24. neo4jService
mongoDBService
productService
Polyglot Setup
13
MATCH (me:Users {id: "' + id + '"})
-[:Friend]-> (:Users)
-[:Bought]-> (p)
return p.id as id
products.find({$or: req.payload})

25. Live Demo
Neo4J + MongoDB
14
https://github.com/mchacki/ms-meets-nosql-examples

26. Multi Model Database
‣ Can natively store several kinds of data models:
‣ Key-value pairs
‣ Documents
‣ Graphs
‣ Delivers query mechanisms for all data models
‣ ACID transactions
‣ Cross collection joins
15

27. Using a Multi-Model Database
16
LoadBalancer
Application Code
Consistency
Joins

28. Using a Multi-Model Database
16
LoadBalancer
Application Code Consistency
Joins

29. Foxx
‣ Customized REST API on top of ArangoDB
‣ Microservice framework
‣ Integrate with other microservices
‣ Reuse your Node.js code and NPM modules
‣ Built-in authentication using OAuth2.0 or HTTP-Basic Auth
‣ Operations are encapsulated in the database
‣ low network traffic, direct data access
‣ increases data privacy
/
(~(
) ) /_/
( _-----_(@ @)
( /
/|/--| V
" " " "
17

30. foxx
productService
Foxx Setup
18
RETURN GRAPH_NEIGHBORS('ecom',
@id, {direction: 'outbound',
minDepth: 2, maxDepth: 2,
neighborCollectionRestriction: 'Products',
includeData: true
})

31. foxx
productService
Foxx Setup
18
RETURN GRAPH_NEIGHBORS('ecom',
@id, {direction: 'outbound',
minDepth: 2, maxDepth: 2,
neighborCollectionRestriction: 'Products',
includeData: true
})

32. foxx
productService
Foxx Setup
18
RETURN GRAPH_NEIGHBORS('ecom',
@id, {direction: 'outbound',
minDepth: 2, maxDepth: 2,
neighborCollectionRestriction: 'Products',
includeData: true
})

33. foxx
productService
Foxx Setup
18
RETURN GRAPH_NEIGHBORS('ecom',
@id, {direction: 'outbound',
minDepth: 2, maxDepth: 2,
neighborCollectionRestriction: 'Products',
includeData: true
})

34. foxx
productService
Foxx Setup
18
RETURN GRAPH_NEIGHBORS('ecom',
@id, {direction: 'outbound',
minDepth: 2, maxDepth: 2,
neighborCollectionRestriction: 'Products',
includeData: true
})

35. Live Demo
Foxx
19
https://github.com/mchacki/ms-meets-nosql-examples

36. ‣ open source and free (Apache 2 license)
‣ sharding & replication
‣ JavaScript throughout (V8 built into server)
‣ drivers for a wide range of languages
‣ web frontend
‣ good & complete documentation
‣ professional as well as community support
20
An overview of other features

37. Thank you
‣ Further questions?
‣ Follow me on twitter/github: @mchacki
‣ Write me a mail: mchacki@arangodb.com
‣ Join or google group: https://groups.google.com/forum/#!forum/arangodb
‣ Visit us at our booth
‣ Free T-shirts
21