Consistent Hashing: Load Balancing in a Changing World

1
Consistent HashingLoad Balancing in a Changing
World

• David Karger, Eric Lehman,
• Tom Leighton, Matt Levine,
• Daniel Lewin, Rina Panigrahy

2
Caches can Load Balance

• Numerous items in central server.
• Requests can swamp server.
• Distribute items among caches.
• Clients get items from caches.
• Server gets only 1 request per item

Server
Items distributed among caches
Users get items from caches
3
Who Caches What?

• Each cache should hold few items
• else cache gets swamped by clients
• Each item should be in few caches
• else server gets swamped by caches
• and cache invalidations/updates expensive
• Browser must know right cache
• fast, local computation

4
A Solution Hashing

• Example y axb (mod n)
• Intuition Assigns items to random caches
• few items per cache
• Easy to compute which cache holds an item

Server
items assigned to caches by hash function.
Users use hash to compute cache for item.
5
Problem Adding Caches

• Suppose a new cache arrives.
• How work it into hash function?
• Natural change
• yaxb (mod n1)
• Problem changes bucket for every item
• every cache will be flushed
• servers get swamped with new requests
• Goal when add bucket, few items move

6
Problem Inconsistent Views

• Each client knows about a different set of
  caches its view
• View affects choice of cache for item
• Same item may hash to many places caches swamp
  server with request for item
• Many items may hash to same place clients swamp
  cache
• Goal despite views, items evenly distributed
  into a few caches each

7
Solution Consistent Hashing

• Use standard hash function to map caches and
  items to points in unit interval.
• random points spread uniformly
• Item assigned to nearest cache in view

Computation easy as standard hash function
8
Properties

• All buckets get roughly same number of items
  (like standard hashing).
• When kth bucket is added only a 1/k fraction of
  items move.
• and only from a few caches

When a cache is added, minimalreshuffling of
cached items is required.
9
Multiple View Properties

• Despite multiple views, each cache gets few items
• no cache overloaded
• Despite multiple views, each item only in few
  caches.
• server protected, cache updates easy

System tolerates multiple, inconsistent views of
caches (also fault tolerant).
10
Load Balancing

• Task distribute items into buckets
• Data to memory locations
• Files to disks
• Tasks to processors
• Web pages to caches (our motivation)
• Goal even distribution

11
Problem No Synchronization

• Each user knows about a different set of caches
  a view
• View affects assignment of items to caches
• Problems when there are multiple views

• The items assigned to a specific cache are
  different in each view.
• These sets could be essentially disjoint for
  standard hash functions.
• Over all views, cache is responsible for too many
  items.
• Cache not large enough to contain active set of
  items

items assigned to one cache over 4 views
12
Multiple Views Cont..

• Item may be assigned to different caches in
  different views.
• Standard hash function may assign item to a
  different cache in every view.
• Result item requested from many caches
• Server swamped with requests for copies of the
  item.
• Hard to update cached copies

View 4
View 3
View 2
View 1
item assigned to different caches in each of 4
views
13
Problem Adding Caches

• New cache means new hash function
• natural change y axb (mod n1)
• Standard hash functions completely redistribute
  items when the range of function changes
• Every cache will be flushed
• server is swamped with requests since items are
  reshuffled between caches.
• Need to broadcast the new hash function to all
  users at the same time
• some kind of global synchronization?...