Ppt On Agreement Protocol In Distributed System

15 Dec

Classification of agreement problems cont… – For all of the problems mentioned above, all non-defective transformers must reach an agreement – For Byzantine and consensual problems, the agreement is fixed on a single value – In the problem of interactive maintenance, the agreement is on a number of common values. In the event of a Bizantin agreement problem, only one processor starts the value in which, as in the other two cases, each processor has its own initial value. AGREEMENT PROTOCOLS Chapter 8. Processes/locations in distributed systems are often competing and cooperating to achieve a common goal. Mutual trust/agreement is in high demand. Introduction – Processes/sites in distributed systems are often competing and cooperating to achieve a common goal. Mutual trust/agreement is in high demand. In distributed databases, data managers may be forced to decide whether to transfer or cancel the transaction – If there is no error, it is easy to reach an agreement. However, in the event of a failure, processes must exchange values with other processes and transmit the values obtained by others several times in order to isolate the effects of a faulty processor. Contract protocols help to reach an agreement in the event of an error. Performance aspects of contractual protocols – The following measures are used: Time: No need for tricks to reach an agreement – Message traffic: the number of messages exchanged to reach an agreement. Memory overload: the amount of information that needs to be stored in processors while the protocol is running.

Classification of contractual problems – There are three known problems of agreement in distributed systems: – Bizantin agreement problem: `Only one value must be agreed. The agreed value is initialized by any processor and all non-defective processors must agree on this value. 2. Consensus problem: Each processor has its own initial value and all non-defective processors must agree on a single common value. 3. Interactive Consistency Problem: Each processor has its own initial value and all non-defective processors must agree on a number of common values. Cont. Applications of contractual algorithms – Tolerant cadence synchronization – Distributed systems require physical clocks to be synchronized – Physical watches have a drift problem – Contractual protocols can help achieve a common cadence value. Atomic Commit in DDBS – DDBS sites need to agree on whether the transaction should be transferred or cancelled – Contract protocols can help reach consensus. Solution for the Bizantin Agreement Problem – First defined and solved by lamport. Source transmits its initial value to all other processors.