Mumbai University > Information Technology > Sem6 > Distributed System

Marks: 10M

Year: May 2015

Client –centric Consistency model-many Inconsistencies can be hidden in cheap way.

• The client-centric consistency models arm at providing a consistent view on a database.
• An important assumption is that concurrent process may be simultaneously updating provide consistency in the face of such concurrency.
• client-centric consistency models are described using the following notations.
• Let xi[t] denote the version of data item x at Li at time t.
• version xi[t] is the result of series of write operations of Li that took place since initialization; we denote this set as WS(xi[t]).

• If operations in WS(xi[t1]) have also been performed at local Lj at later time t2, we write WS(xi[t1];xj[t2]).

  The Client-Centric consistency models are:

  1) Monotonic Reads

  2) Monotonic Writes

  3) Read your Writes

  4) Writes follow Reads

  5) Eventual Consistency

1) Monotonic reads :

A data store is said to provide monotonic read consistency if the following conditions helds.

• If a process reads the value of a algorithm x by what process will always return that same value or more recent value.
• In other words, monotonic read consistency guarantees that if a process has seen a value of x at a more time t, it will never see an older version of x at a later time.

2) Monotonic writes: in a monotonic write consistent store, the following conditions holds:

• A write operation by a process on a data item x is completed before any successive write operation on x by the same process.

3) Read your writes:

A database is said to provide read your write consistency, if the following condition holds:

• The effect of a write operation by a process on data item x will always be seen by a successive read operation on x by the same process.

4) Write your reads:

A data store is said to provide writes follows reads consistency, if the following conditions holds:

• A write operation by a process on a data item x following a previous read operation on x by the same process is generated to take place on the same or a more recent value of x that was read.

5) Eventual Consistency:

• Data stores there are eventually consistent have properly that in absence of updates, all replicas coverage toward identical copies of each other.
• Eventual consistency essentially requires only that updates are guaranteed to propagate to all replicas.