Sequential Consistency Model and Linearizability Model in Distributed System(Base Models).Sequential Consistency Model In Distributed System Consistency Models In Distributed Systems Causal Consistency In Distributed System Consistency Models In Distributed Systems Video Lecture. Implementing Sequential Consistency Model Linearizability Vs. Sequential Consistency Examples Distributed Systems Replication And Consistency Distributed Systems.<\/p>\n
[fvplayer id=”71″]<\/p>\n
In distributed systems<\/strong>, maintaining consistency<\/strong> is a crucial challenge. Two important consistency models used to define how operations should be executed and observed are Sequential Consistency<\/strong> and Linearizability<\/strong>. These models help in designing reliable and predictable systems, ensuring correct execution of concurrent operations<\/strong>.<\/p>\n Sequential Consistency<\/strong> ensures that the result of execution is as if<\/strong> all operations were executed in some sequential order<\/strong>, and each process sees operations in the same order<\/strong>. However, it does not<\/strong> guarantee real-time execution order.<\/p>\n Key Property:<\/strong> Consider two processes ( Possible Sequentially Consistent Execution Orders:<\/strong> Even though execution may not match real-time order<\/strong>, all processes see changes in a consistent sequence<\/strong>.<\/p>\n Linearizability (Atomic Consistency)<\/strong> is a stronger consistency model<\/strong> than Sequential Consistency<\/strong>. It ensures that each operation appears to take effect instantaneously at some point between its start and end time<\/strong>.<\/p>\n Key Property:<\/strong> Consider the same shared variable If Sequential Consistency:<\/strong> Linearizability:<\/strong> Sequential Consistency<\/strong> ensures operations appear in some global sequence<\/strong>, but real-time execution may be violated. \u00a0Which consistency model do you think is more suitable for real-world distributed systems?<\/strong><\/p>\n Here\u2019s a detailed and clear comparison of the Sequential Consistency Model<\/strong> and the Linearizability Model<\/strong> in distributed systems \u2014 two important consistency models<\/strong> used to reason about the behavior of concurrent systems like databases, caches, and shared memory.<\/p>\n In distributed systems, consistency models<\/strong> define the rules for the visibility and ordering of read\/write operations<\/strong> across multiple nodes or processes.<\/p>\n The two commonly discussed models are:<\/p>\n A system is sequentially consistent<\/strong> if the result of execution is the same as if<\/strong>:<\/p>\n All operations were executed in some sequential order<\/strong><\/p>\n<\/li>\n The order of operations of each individual process is preserved<\/strong><\/p>\n<\/li>\n<\/ul>\n Let\u2019s say two processes perform the following writes:<\/p>\n\u00a0What is Sequential Consistency?<\/strong><\/h3>\n
\u00a0Definition<\/strong><\/h3>\n
Operations from different processes may be interleaved, but they must respect the program order within each process.<\/em><\/p>\n\u00a0Example of Sequential Consistency<\/strong><\/h3>\n
P1<\/code> and P2<\/code>) updating a shared variable X<\/code>, initially set to 0<\/code>.<\/p>\nOperations:<\/strong><\/h4>\n
P1: X = 1<\/code>P2: print(X)<\/code> (can see X = 0<\/code> or X = 1<\/code>, depending on execution order)<\/p>\n
Order 1:<\/strong> P1 \u2192 P2<\/code> \u2192 (P2 sees X = 1<\/code>)
Order 2:<\/strong> P2 \u2192 P1<\/code> \u2192 (P2 sees X = 0<\/code>)<\/p>\n\u00a0What is Linearizability?<\/strong><\/h3>\n
\u00a0Definition<\/strong><\/h3>\n
All operations must appear to execute in a real-time order.<\/em> If operation A<\/code> completes before operation B<\/code> starts, then A<\/code> must be observed before B<\/code>.<\/p>\n\u00a0Example of Linearizability<\/strong><\/h3>\n
X<\/code> with processes P1<\/code> and P2<\/code>.<\/p>\nOperations:<\/strong><\/h4>\n
P1: X = 1<\/code>P2: print(X)<\/code><\/p>\nP2<\/code> starts reading after P1<\/code> has updated X<\/code>, P2 must see X = 1<\/code><\/strong>.
Linearizability ensures that no process ever sees an outdated value<\/strong> after an update.<\/p>\n\u00a0Difference Between Sequential Consistency and Linearizability<\/strong><\/h3>\n
\n\n
\n Feature<\/th>\n Sequential Consistency<\/strong><\/th>\n Linearizability<\/strong><\/th>\n<\/tr>\n<\/thead>\n\n \n Ordering<\/strong><\/td>\n Preserves execution order but allows reordering across processes<\/td>\n Preserves both execution order and real-time constraints<\/td>\n<\/tr>\n \n Real-Time Guarantee<\/strong><\/td>\n No<\/td>\n Yes<\/td>\n<\/tr>\n \n Use Case<\/strong><\/td>\n Useful in systems where exact real-time order is not required<\/td>\n Required in financial transactions, database consistency, and real-time systems<\/td>\n<\/tr>\n \n Performance<\/strong><\/td>\n Better performance as it allows reordering<\/td>\n Expensive as it requires strict order maintenance<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n \u00a0Real-World Use Cases<\/strong><\/h3>\n
Cloud Storage (Amazon S3, Google Drive)<\/strong> \u2013 Ensures eventual order but does not guarantee real-time updates.
Distributed Caching<\/strong> \u2013 Helps in optimizing performance by allowing relaxed ordering.<\/p>\n
Bank Transactions (Atomicity in Databases)<\/strong> \u2013 Ensures that withdrawals and deposits reflect in real-time.
Leader Election (Raft, Paxos)<\/strong> \u2013 Ensures strict order in distributed consensus.<\/p>\n\u00a0Summary<\/strong><\/h3>\n
Linearizability<\/strong> is a stronger<\/strong> guarantee where operations must appear instantaneous and respect real-time order<\/strong>.
Linearizability is expensive<\/strong> but necessary for critical systems like banking and distributed databases.<\/p>\n
\n\ud83e\udde0 Consistency Models in Distributed Systems<\/strong><\/h1>\n
\n\ud83d\udd39 1. Sequential Consistency Model<\/strong><\/h2>\n
\ud83d\udcd8 Definition:<\/h3>\n
\n
\ud83e\uddea Example:<\/h3>\n