第 18 章：复制（REPLICATION）¶

Replication – maintenance of copies of data at multiple computers:

Motivations¶

Motivations for replication include:

replication of immutable data is trivial:it increases performance with little cost to the system
replication of changing data incurs overheads in the form of protocols degined to ensure that clients receive up-to-date data: this limites to the effectiveness of replication as a performance-enhancement technique.

Apart from delays due to pessimistic concurrency control conflicts (due to data locking), the factors that are relevant to high availability are:

server failures
- replication is a technique for automatically maintaining the availability of data despite server failures
- if data are replicated at two or more failure-independent servers, then client sofrware may be able to access data at an alternative server when the default server fail or become unreachable.
- the percentage of time during which the SERVICE is available can be enhanced by replicating server data
- if each of n servers has an independent probability p of failing or becoming unreachable, then the availability of an object stored at each of these servers is : $1 - p^n$
network partitions and disconnected operation
- user need to working in a disconnected environment
- often prepare by copying heavily used data, such as the contents of a shared diary, from their usual environment to the laptop or mobile phone
- when the user consults or updates the data, they risk reading data that someone else has altered in the meantime
- disconnected working is ONLY feasible if the user (or the application on the user’s behalf) can cope with stale data and can later reslove any conflicts that arise

high available data is NOT necessarily strictly correct data, it may be out of date, incomplete, or have some kind of error, etc.
a fault-tolerant service, by contrast, ALWAYS guarantees strictly correct behaviour despite a certain number and type of faults.
the correctness concerns :
- the freshness of data supplied to the client, and
- the effects of the client’s operations upon the data,
- sometimes also concerns the timeliness of the service’s responses.
the same basic technique used for high availability – that of replicating data and functionality between computes – is also applicable for achieving fault tolerance:
- more replication server, higher availability
- but system must manage the coordination of its components precisely to maintain the correctness guarantees in the face of failures, which may occur at any time.

replication transparency:
- clients should not normally have to be aware that MULTIPLE PHYSICAL copies of data exists.
- as far as clients are concerned, data are organized as INDIVIDUAL LOGICAL objects and they identify ONLY ONE item in each case when they request an operation to be performed.
- clients expect operations to return ONLY ONE set of values.
consistency (can be vary in strength between applications): concerns wherther the operations performed upon a collection of replicaed objects produce result that meet the spcification of correctness for those objects.