2020-03-10 08:38:48 +07:00
# Design Amazon's sales rank by category feature
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
*Note: This document links directly to relevant areas found in the [system design topics ](https://github.com/donnemartin/system-design-primer#index-of-system-design-topics ) to avoid duplication. Refer to the linked content for general talking points, tradeoffs, and alternatives.*
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
## Step 1: Outline use cases and constraints
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
> Gather requirements and scope the problem.
> Ask questions to clarify use cases and constraints.
> Discuss assumptions.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
Without an interviewer to address clarifying questions, we'll define some use cases and constraints.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Use cases
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
#### We'll scope the problem to handle only the following use case
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* **Service** calculates the past week's most popular products by category
* **User** views the past week's most popular products by category
* **Service** has high availability
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
#### Out of scope
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* The general e-commerce site
* Design components only for calculating sales rank
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Constraints and assumptions
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
#### State assumptions
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* Traffic is not evenly distributed
* Items can be in multiple categories
* Items cannot change categories
* There are no subcategories ie `foo/bar/baz`
* Results must be updated hourly
* More popular products might need to be updated more frequently
* 10 million products
* 1000 categories
* 1 billion transactions per month
* 100 billion read requests per month
* 100:1 read to write ratio
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
#### Calculate usage
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
**Clarify with your interviewer if you should run back-of-the-envelope usage calculations.**
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* Size per transaction:
* `created_at` - 5 bytes
* `product_id` - 8 bytes
* `category_id` - 4 bytes
* `seller_id` - 8 bytes
* `buyer_id` - 8 bytes
* `quantity` - 4 bytes
* `total_price` - 5 bytes
* Total: ~40 bytes
* 40 GB of new transaction content per month
* 40 bytes per transaction * 1 billion transactions per month
* 1.44 TB of new transaction content in 3 years
* Assume most are new transactions instead of updates to existing ones
* 400 transactions per second on average
* 40,000 read requests per second on average
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
Handy conversion guide:
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* 2.5 million seconds per month
* 1 request per second = 2.5 million requests per month
* 40 requests per second = 100 million requests per month
* 400 requests per second = 1 billion requests per month
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
## Step 2: Create a high level design
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
> Outline a high level design with all important components.
2017-03-05 12:06:17 +07:00
![Imgur ](http://i.imgur.com/vwMa1Qu.png )
2020-03-10 08:38:48 +07:00
## Step 3: Design core components
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
> Dive into details for each core component.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Use case: Service calculates the past week's most popular products by category
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We could store the raw **Sales API** server log files on a managed **Object Store** such as Amazon S3, rather than managing our own distributed file system.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
**Clarify with your interviewer how much code you are expected to write**.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We'll assume this is a sample log entry, tab delimited:
2017-03-05 12:06:17 +07:00
```
timestamp product_id category_id qty total_price seller_id buyer_id
t1 product1 category1 2 20.00 1 1
t2 product1 category2 2 20.00 2 2
t2 product1 category2 1 10.00 2 3
t3 product2 category1 3 7.00 3 4
t4 product3 category2 7 2.00 4 5
t5 product4 category1 1 5.00 5 6
...
```
2020-03-10 08:38:48 +07:00
The **Sales Rank Service** could use **MapReduce** , using the **Sales API** server log files as input and writing the results to an aggregate table `sales_rank` in a **SQL Database** . We should discuss the [use cases and tradeoffs between choosing SQL or NoSQL ](https://github.com/donnemartin/system-design-primer#sql-or-nosql ).
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We'll use a multi-step **MapReduce** :
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* **Step 1** - Transform the data to `(category, product_id), sum(quantity)`
* **Step 2** - Perform a distributed sort
2017-03-05 12:06:17 +07:00
2019-05-07 17:24:41 +07:00
```python
2017-03-05 12:06:17 +07:00
class SalesRanker(MRJob):
def within_past_week(self, timestamp):
2020-03-10 08:38:48 +07:00
"""Return True if timestamp is within past week, False otherwise."""
2017-03-05 12:06:17 +07:00
...
def mapper(self, _ line):
2020-03-10 08:38:48 +07:00
"""Parse each log line, extract and transform relevant lines.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
Emit key value pairs of the form:
2017-03-05 12:06:17 +07:00
(category1, product1), 2
(category2, product1), 2
(category2, product1), 1
(category1, product2), 3
(category2, product3), 7
(category1, product4), 1
"""
timestamp, product_id, category_id, quantity, total_price, seller_id, \
buyer_id = line.split('\t')
if self.within_past_week(timestamp):
yield (category_id, product_id), quantity
def reducer(self, key, value):
2020-03-10 08:38:48 +07:00
"""Sum values for each key.
2017-03-05 12:06:17 +07:00
(category1, product1), 2
(category2, product1), 3
(category1, product2), 3
(category2, product3), 7
(category1, product4), 1
"""
yield key, sum(values)
def mapper_sort(self, key, value):
2020-03-10 08:38:48 +07:00
"""Construct key to ensure proper sorting.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
Transform key and value to the form:
2017-03-05 12:06:17 +07:00
(category1, 2), product1
(category2, 3), product1
(category1, 3), product2
(category2, 7), product3
(category1, 1), product4
2020-03-10 08:38:48 +07:00
The shuffle/sort step of MapReduce will then do a
distributed sort on the keys, resulting in:
2017-03-05 12:06:17 +07:00
(category1, 1), product4
(category1, 2), product1
(category1, 3), product2
(category2, 3), product1
(category2, 7), product3
"""
category_id, product_id = key
quantity = value
yield (category_id, quantity), product_id
def reducer_identity(self, key, value):
yield key, value
def steps(self):
2020-03-10 08:38:48 +07:00
"""Run the map and reduce steps."""
2017-03-05 12:06:17 +07:00
return [
self.mr(mapper=self.mapper,
reducer=self.reducer),
self.mr(mapper=self.mapper_sort,
reducer=self.reducer_identity),
]
```
2020-03-10 08:38:48 +07:00
The result would be the following sorted list, which we could insert into the `sales_rank` table:
2017-03-05 12:06:17 +07:00
```
(category1, 1), product4
(category1, 2), product1
(category1, 3), product2
(category2, 3), product1
(category2, 7), product3
```
2020-03-10 08:38:48 +07:00
The `sales_rank` table could have the following structure:
2017-03-05 12:06:17 +07:00
```
id int NOT NULL AUTO_INCREMENT
category_id int NOT NULL
total_sold int NOT NULL
product_id int NOT NULL
PRIMARY KEY(id)
FOREIGN KEY(category_id) REFERENCES Categories(id)
FOREIGN KEY(product_id) REFERENCES Products(id)
```
2020-03-10 08:38:48 +07:00
We'll create an [index ](https://github.com/donnemartin/system-design-primer#use-good-indices ) on `id ` , `category_id` , and `product_id` to speed up lookups (log-time instead of scanning the entire table) and to keep the data in memory. Reading 1 MB sequentially from memory takes about 250 microseconds, while reading from SSD takes 4x and from disk takes 80x longer.< sup >< a href = https://github.com/donnemartin/system-design-primer#latency-numbers-every-programmer-should-know > 1</ a ></ sup >
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Use case: User views the past week's most popular products by category
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* The **Client** sends a request to the **Web Server** , running as a [reverse proxy ](https://github.com/donnemartin/system-design-primer#reverse-proxy-web-server )
* The **Web Server** forwards the request to the **Read API** server
* The **Read API** server reads from the **SQL Database** `sales_rank` table
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We'll use a public [**REST API** ](https://github.com/donnemartin/system-design-primer#representational-state-transfer-rest ):
2017-03-05 12:06:17 +07:00
```
$ curl https://amazon.com/api/v1/popular?category_id=1234
```
2020-03-10 08:38:48 +07:00
Response:
2017-03-05 12:06:17 +07:00
```
{
"id": "100",
"category_id": "1234",
"total_sold": "100000",
"product_id": "50",
},
{
"id": "53",
"category_id": "1234",
"total_sold": "90000",
"product_id": "200",
},
{
"id": "75",
"category_id": "1234",
"total_sold": "80000",
"product_id": "3",
},
```
2020-03-10 08:38:48 +07:00
For internal communications, we could use [Remote Procedure Calls ](https://github.com/donnemartin/system-design-primer#remote-procedure-call-rpc ).
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
## Step 4: Scale the design
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
> Identify and address bottlenecks, given the constraints.
2017-03-05 12:06:17 +07:00
![Imgur ](http://i.imgur.com/MzExP06.png )
2020-03-10 08:38:48 +07:00
**Important: Do not simply jump right into the final design from the initial design!**
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
State you would 1) **Benchmark/Load Test** , 2) **Profile** for bottlenecks 3) address bottlenecks while evaluating alternatives and trade-offs, and 4) repeat. See [Design a system that scales to millions of users on AWS ](../scaling_aws/README.md ) as a sample on how to iteratively scale the initial design.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
It's important to discuss what bottlenecks you might encounter with the initial design and how you might address each of them. For example, what issues are addressed by adding a **Load Balancer** with multiple **Web Servers** ? **CDN** ? **Master-Slave Replicas** ? What are the alternatives and **Trade-Offs** for each?
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We'll introduce some components to complete the design and to address scalability issues. Internal load balancers are not shown to reduce clutter.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
*To avoid repeating discussions*, refer to the following [system design topics ](https://github.com/donnemartin/system-design-primer#index-of-system-design-topics ) for main talking points, tradeoffs, and alternatives:
2017-03-09 10:20:23 +07:00
2020-03-10 08:38:48 +07:00
* [DNS ](https://github.com/donnemartin/system-design-primer#domain-name-system )
* [CDN ](https://github.com/donnemartin/system-design-primer#content-delivery-network )
* [Load balancer ](https://github.com/donnemartin/system-design-primer#load-balancer )
* [Horizontal scaling ](https://github.com/donnemartin/system-design-primer#horizontal-scaling )
* [Web server (reverse proxy) ](https://github.com/donnemartin/system-design-primer#reverse-proxy-web-server )
* [API server (application layer) ](https://github.com/donnemartin/system-design-primer#application-layer )
* [Cache ](https://github.com/donnemartin/system-design-primer#cache )
* [Relational database management system (RDBMS) ](https://github.com/donnemartin/system-design-primer#relational-database-management-system-rdbms )
* [SQL write master-slave failover ](https://github.com/donnemartin/system-design-primer#fail-over )
* [Master-slave replication ](https://github.com/donnemartin/system-design-primer#master-slave-replication )
* [Consistency patterns ](https://github.com/donnemartin/system-design-primer#consistency-patterns )
* [Availability patterns ](https://github.com/donnemartin/system-design-primer#availability-patterns )
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
The **Analytics Database** could use a data warehousing solution such as Amazon Redshift or Google BigQuery.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We might only want to store a limited time period of data in the database, while storing the rest in a data warehouse or in an **Object Store** . An **Object Store** such as Amazon S3 can comfortably handle the constraint of 40 GB of new content per month.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
To address the 40,000 *average* read requests per second (higher at peak), traffic for popular content (and their sales rank) should be handled by the **Memory Cache** instead of the database. The **Memory Cache** is also useful for handling the unevenly distributed traffic and traffic spikes. With the large volume of reads, the **SQL Read Replicas** might not be able to handle the cache misses. We'll probably need to employ additional SQL scaling patterns.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
400 *average* writes per second (higher at peak) might be tough for a single **SQL Write Master-Slave** , also pointing to a need for additional scaling techniques.
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
SQL scaling patterns include:
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* [Federation ](https://github.com/donnemartin/system-design-primer#federation )
* [Sharding ](https://github.com/donnemartin/system-design-primer#sharding )
* [Denormalization ](https://github.com/donnemartin/system-design-primer#denormalization )
* [SQL Tuning ](https://github.com/donnemartin/system-design-primer#sql-tuning )
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
We should also consider moving some data to a **NoSQL Database** .
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
## Additional talking points
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
> Additional topics to dive into, depending on the problem scope and time remaining.
2017-03-05 12:06:17 +07:00
#### NoSQL
2020-03-10 08:38:48 +07:00
* [Key-value store ](https://github.com/donnemartin/system-design-primer#key-value-store )
* [Document store ](https://github.com/donnemartin/system-design-primer#document-store )
* [Wide column store ](https://github.com/donnemartin/system-design-primer#wide-column-store )
* [Graph database ](https://github.com/donnemartin/system-design-primer#graph-database )
* [SQL vs NoSQL ](https://github.com/donnemartin/system-design-primer#sql-or-nosql )
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Caching
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* Where to cache
* [Client caching ](https://github.com/donnemartin/system-design-primer#client-caching )
* [CDN caching ](https://github.com/donnemartin/system-design-primer#cdn-caching )
* [Web server caching ](https://github.com/donnemartin/system-design-primer#web-server-caching )
* [Database caching ](https://github.com/donnemartin/system-design-primer#database-caching )
* [Application caching ](https://github.com/donnemartin/system-design-primer#application-caching )
* What to cache
* [Caching at the database query level ](https://github.com/donnemartin/system-design-primer#caching-at-the-database-query-level )
* [Caching at the object level ](https://github.com/donnemartin/system-design-primer#caching-at-the-object-level )
* When to update the cache
* [Cache-aside ](https://github.com/donnemartin/system-design-primer#cache-aside )
* [Write-through ](https://github.com/donnemartin/system-design-primer#write-through )
* [Write-behind (write-back) ](https://github.com/donnemartin/system-design-primer#write-behind-write-back )
* [Refresh ahead ](https://github.com/donnemartin/system-design-primer#refresh-ahead )
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Asynchronism and microservices
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* [Message queues ](https://github.com/donnemartin/system-design-primer#message-queues )
* [Task queues ](https://github.com/donnemartin/system-design-primer#task-queues )
* [Back pressure ](https://github.com/donnemartin/system-design-primer#back-pressure )
* [Microservices ](https://github.com/donnemartin/system-design-primer#microservices )
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Communications
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* Discuss tradeoffs:
* External communication with clients - [HTTP APIs following REST ](https://github.com/donnemartin/system-design-primer#representational-state-transfer-rest )
* Internal communications - [RPC ](https://github.com/donnemartin/system-design-primer#remote-procedure-call-rpc )
* [Service discovery ](https://github.com/donnemartin/system-design-primer#service-discovery )
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Security
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
Refer to the [security section ](https://github.com/donnemartin/system-design-primer#security ).
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Latency numbers
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
See [Latency numbers every programmer should know ](https://github.com/donnemartin/system-design-primer#latency-numbers-every-programmer-should-know ).
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
### Ongoing
2017-03-05 12:06:17 +07:00
2020-03-10 08:38:48 +07:00
* Continue benchmarking and monitoring your system to address bottlenecks as they come up
* Scaling is an iterative process