1-introduction Flashcards

(40 cards)

1
Q

What is a distributed system?

A

A collection of autonomous computing elements that appears to its users as a single coherent system.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the two key characteristics of nodes in a DS?

A
  1. They need to communicate with each other. 2. There is no global clock - each node has its own time.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is location transparency?

A

Users cannot tell where computation happens or where data is stored.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is replication transparency?

A

Users cannot tell whether data is replicated across multiple nodes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the difference between distributed and decentralised?

A

Decentralised systems have more redundant peer-to-peer connections. Distributed focuses on appearing as one coherent system.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the Integrative view of building DS?

A

Adding a networked computer into a larger system.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the Expansive view of building DS?

A

Extending one computer with more computers.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the Architecture challenge in DS?

A

How do we organise? Client-server or peer-to-peer? Microservices? Layered architecture?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the Process challenge in DS?

A

What kind of processes - threads

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the Communication challenge in DS?

A

RPC? API? Message queues? What protocols and middleware?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the Coordination challenge in DS?

A

Synchronising actions of all the computers in the system.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the Naming challenge in DS?

A

How do we identify resources?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the Consistency challenge in DS?

A

We need data copies and we need to synchronise those copies across nodes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the Fault Tolerance challenge in DS?

A

Parts will fail - detect and fix without taking down the whole system.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the Security challenge in DS?

A

Need authentication and encrypted communication.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the Latency challenge in DS?

A

Need to minimise and hide network delay.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is the Processing Power challenge in DS?

A

Need to allocate enough processing power across nodes.

18
Q

What is the Storage challenge in DS?

A

How do we distribute storage across the system?

19
Q

Name 4 examples of distributed systems.

A

Internet, WWW, Cell networks, Cloud.

20
Q

What is parallel computing?

A

Multiprocessor/core vs multicomputer systems that can share memory or have private memory.

21
Q

What was Distributed Shared Memory (DSM)?

A

Implementing shared memory model on multicomputer using virtual memory techniques. Failed due to lower performance than true multiprocessors, now abandoned.

22
Q

How did DSM work?

A

Map all main memory pages to a single address space. OS traps and fetches pages from correct computer.

23
Q

What is Cluster Computing?

A

High-end systems connected locally (LAN). Same OS and hardware (homogeneous). Single program manages entire cluster.

24
Q

What is Grid Computing?

A

Lots of nodes anywhere. Flexible, secure, coordinated resource sharing. Virtual organisations share resources for common goals.

25
What is the difference between Cluster and Grid computing?
Cluster: local, homogeneous, single admin. Grid: distributed, heterogeneous, multiple organisations.
26
What are the 4 layers of Cloud Computing?
Hardware (unseen by users), Infrastructure (virtualisation), Platform (development/deployment), Application (cloud apps like office suites).
27
What is Enterprise Application Integration (EAI)?
Allowing direct communication between applications in an organisation rather than having a client collate results.
28
What does ACID stand for?
Atomic, Consistent, Isolated, Durable.
29
What is Atomicity in ACID?
Transaction completes fully or not at all - never partial.
30
What is Consistency in ACID?
Transaction brings database from one valid state to another.
31
What is Isolation in ACID?
Concurrent transactions don't interfere with each other.
32
What is Durability in ACID?
Committed transactions persist even after system failure.
33
What is a Transaction Processing Monitor (TPM)?
Receives request, breaks into operations for different servers, sends and collects responses. All succeed → commit. Any fail → abort and rollback.
34
What is RPC (Remote Procedure Call)?
Application calls function, RPC packages call and parameters into message, sends to remote server, result packaged and returned. Abstracted from programmer.
35
What is MOM (Message Oriented Middleware)?
Asynchronous messaging - application publishes to queue, MOM holds it, subscribed apps receive and process.
36
What is Ubiquitous Computing?
Embedded systems in your environment that operate autonomously, unobtrusively, without user interaction. System knows who/where/what you are.
37
What is Mobile Computing?
Devices that move between different local services. No guaranteed connectivity or stable routing. Must discover what's nearby.
38
What are the two approaches to mobile computing?
Cloud only (thin client, far away, can be slow) or Cloud with Edge (smaller server closer to client, some local processing).
39
What are Sensor Networks?
Thousands of sensors on small battery-powered nodes with some memory and compute. Process locally or send to main database.
40
What is the Cloud Edge Continuum?
Trade-off: closer to cloud = more reliable/compute but less mobile/responsive. Closer to edge = opposite.