Caching

Question 1

Why is caching important?

Answer 1

it helps make better use of the resources you have, and makes calls faster

Question 2

Locality of Reference Principle

Answer 2

recently requested data is likely to be requested again

Question 3

Where is a cache usually found?

Answer 3

closest to the front end to return data quickly without using downstream resources

Question 4

What is an application server cache?

Answer 4

when the cache is placed on the request layer node itself, the node quickly returns local cached data if it exists, otherwise it queries from disk

Question 5

What are the benefits of using an application server cache?

Answer 5

caches can be in memory, which is super fast, or on a node’s local disk, which is faster than network storage

Question 6

What are some tradeoffs of using an application server cache?

Answer 6

it performs worse the more the load balancer distributes the load (ie as there are more nodes, it gets worse)

Question 7

What is a distributed cache?

Answer 7

each node owns part of the cached data, which is divided up using a hashing function

Question 8

What are the benefits of using a distributed cache?

Answer 8

you can easily increase cache space by adding nodes to the pool

Question 9

What are some tradeoffs of using a distributed cache?

Answer 9

if a node is missing, there is an automatic cache miss

Question 10

How could you resolve an issue with a missing node in a distributed cache?

Answer 10

technically you can replicate the information, but that becomes complicated quickly, when really the outcome is that the request just has to pull from the origin

Question 11

What is a global cache?

Answer 11

all nodes use a single cache space, treats cache like a local one

Question 12

What does a global cache require?

Answer 12

a server or file store that is faster than the original store and accessible by all request layer nodes

Question 13

What are some benefits of using a global cache?

Answer 13

it’s very effective in some architectures, like ones with specialized hardware or a fixed dataset that needs to be cached

Question 14

What are some tradeoffs of using a global cache?

Answer 14

it can get complicated as the number of clients and requests increase

Question 15

What are the two forms of a global cache?

Answer 15

1. when there is a cache miss, the cache is responsible for retrieving the missing piece of data from the store (used the majority of the time)
2. when there is a cache miss, the request node is responsible for retrieving the missing data (used in special cases, like when the files are large)

Question 16

What is a CDN?

Answer 16

content distribution network, used for serving large amounts of static media

Question 17

What is cache invalidation?

Answer 17

when data is modified in the database, it needs to be marked as invalid in the cache

Question 18

What is a write-through cache?

Answer 18

data is written into the cache and the database at the same time

Question 19

What are some benefits of a write-through cache?

Answer 19

allows for fast retrieval and loses nothing if the system is disrupted

Question 20

What are some tradeoffs of a write-through cache?

Answer 20

there is higher latency for write operations

Question 21

What is a write-around cache?

Answer 21

data is written directly to permanent storage, bypassing the cache

Question 22

What are some tradeoffs of a write-around cache?

Answer 22

a read request for recently written data will result in a cache miss

Question 23

What is a write-back cache?

Answer 23

data is only written to the cache, and completion is confirmed immediately to the client. write to permanent storage is done in batches

Question 24

What are some benefits to a write-back cache?

Answer 24

low latency and high throughput

Question 25

What are some tradeoffs of a write-back cache?

Answer 25

risk of data loss if the system is disrupted

Question 26

What is FIFO?

Answer 26

First In, First Out, evicts first block accessed

Question 27

What is LIFO?

Answer 27

Last in, First out, evicts the block accessed most recently

Question 28

What is LRU?

Answer 28

Least Recently Used, discards least recently used items first

Question 29

What is MRU?

Answer 29

Most Recently Used, discards most recently used items first

Question 30

What is LFU?

Answer 30

Least Frequently Used, counts how often an item is needed and discards least used first

Question 31

What is RR?

Answer 31

Random Replacement, randomly select an item to discard when space is needed