Week 3 - Spreadsheets and Storage

Question 1

What type of programming are spreadsheets a form of?

Answer 1

Functional programming

Question 2

Give 2 use cases of spreadsheets.

Answer 2

Reading a cell
Updating a cell

Question 3

How else can a spreadsheet be seen?

Answer 3

As a store of documents

Question 4

What does the repository design pattern encapsulate?

Answer 4

A set of objects persisted in a data store and the operations performed over them. It provides a clean separation between the domain and mapping layers.

Question 5

How can an enterprise computer system store data?

Answer 5

Using a centralised or decentralised/distributed database

Question 6

What is an enterprise using a monolith likely to store its data in?

Answer 6

A centralised database

Question 7

What are some pros of using centralised database organisation?

Answer 7

It is easy to ensure that the accuracy, completeness, and consistency of data is maintained

Question 8

What are the two things a transaction can do?

Answer 8

Commit (completes successfully), and the database is moved to a new consistent state.

Aborts (completes unsuccessfully), and the database is restored to its previous consistent state.

Question 9

Transactions have the ACID properties. What does ACID stand for?

Answer 9

Atomicity, Consistency, Isolation, Durability

Question 10

What does atomicity mean in terms of centralised database transactions?

Answer 10

All or nothing - transactions either happen or they don’t

Question 11

What does consistency mean in terms of centralised database transactions?

Answer 11

Before a transaction the database is in a consistent state, and after the transaction the database is in a consistent state

Question 12

What does isolation mean in terms of centralised database transactions?

Answer 12

Serialisability - transactions cannot view other in-progress transactions

Question 13

What does durability mean in terms of centralised database transactions?

Answer 13

If a transaction takes place, its effects will be seen

Question 14

What is an enterprise using microservices likely to store its data in?

Answer 14

A distributed/decentralised database

Question 15

What are some cons of distributed databases?

Answer 15

It is difficult to ensure the accuracy, completeness and consistency of data is maintained

Question 16

How can a distributed database be managed?

Answer 16

With two-phase commit or sagas

Question 17

Describe a two-phase commit.

Answer 17

A coordinator transaction asks a number of participant transactions to vote on whether they are prepared to commit a change.

Each participant holds locks on its data involved in the transaction until the coordinator decides to commit or abort.

If all participants vote to commit, then the coordinator instructs all participants to commit. Otherwise, if any participant votes to abort or times out, the coordinator instructs all participants to abort.

Question 18

What is a problem with a two-phase commit?

Answer 18

The locks on the transaction data means other transactions can’t take place at the same time.

Question 19

Describe a saga.

Answer 19

A coordinator transaction asks each participant to commit or abort in sequence.

Each participant holds a lock on its data involved in the transaction only until it decides to commit or abort.

Question 20

What happens in a saga if a participant aborts?

Answer 20

The sequence ends immediately and compensating transactions are made to undo the work of those participants that have already committed - a saga sacrifices atomicity and relies on eventual consistency.

Question 21

Among the desirable properties of a distributed system are CAP - what does this stand for?

Answer 21

Consistency, Availability, Partition-tolerance

Question 22

What does consistency mean in the CAP Theorem?

Answer 22

The same response is received from any node/microservice

Question 23

What does availability mean in the CAP Theorem?

Answer 23

A response is received from every node/microservice

Question 24

What does partition-tolerance mean in the CAP Theorem?

Answer 24

The system works even if nodes/microservices become disconnected

Question 25

What does Eric Brewer's CAP Theorem say?

Answer 25

Although there are three fundamental guarantees (CAP) that a distributed system could make, one can only ever have two of them at the time.

Question 26

What are the three ways in which you can decompose a database and what do they mean?

Answer 26

Split the database then the code - this change is unlikely to deliver much short-term benefit as the monolith remains, and transaction atomicity is lost. Split the code then the database - this change will hopefully deliver immediate short-term benefit from the new service, and the confidence to complete the decomposition. Split the code and the database - this is a much bigger change, and the benefit will be long-term as there is so much work to do, so this should be avoided.