Que4.27. What are infinitesimal commit protocols? Answer 1. infinitesimal commit protocols are the crucial element in supporting global atomicity of distributed deals. 2. Two- phase commit protocol( 2PC) is the standard infinitesimal commit protocol. 3. 2PC is important to guarantee correctness parcels in the complex distributed world whilst at the same time it reduces community due to high fragment and communication outflow and locking during windows of vulnerability. 4. An infinitesimal commitment problem requires processes to agree on a common outgrowth which can be either commit or repeal. 5. An infinitesimal commit protocol must guarantee the following infinitesimal commitment parcels AC1 All processes that reach an outgrowth reach the same bone
. AC2 A process can not reverse its outgrowth after it has reached one. AC3 The commit outgrowth can only be reached if all party suggested Yes. AC4 If there are no failures and all actors suggested Yes, also the outgrowth will be commit. AC5 Consider any prosecution containing only failures that the protocol is designed to tolerate. 6 At any point in the prosecution, if all being failures are repaired and no new failures do for sufficiently long, also all processes will ultimately reach an outgrowth. Que4.28. Explain replication and its types in distributed system. Answer 1. Replication is a fashion of replicating data over a system. 2. Replication is a key to the effectiveness of distributed systems in that, it provides enhanced performance, high vacuity and high fault forbearance. 3. The replication is the conservation of clones of data at multiple computers. 4. Replication is a fashion for enhancing a service. 5. When data are replicated, the replication translucency is needed i.e., guests shouldn’t typically have to be apprehensive that multiple clones of data live. Types of replication Active replication 1. In active replication each customer request is reused by all the waiters. 2. This requires that the process hosted by the waiters is deterministic, i.e., given the same original state and a request sequence, all processes will produce the same response sequence and end up in the same final state. 3. In order to make all the waiters admit the same sequence of operations, an infinitesimal broadcast protocol must be used. 4. An infinitesimal broadcast protocol guarantees that either all the waiters admit a communication or none, plus that they all admit dispatches in the same order. ii. Passive replication 1. In unresistant replication there’s only one garçon( called primary) that processes customer requests. 2. After recycling a request, the primary garçon updates the state on the other( backup) waiters and sends back the response to the customer. 3. still, one of the backup waiters takes its place, If the primary garçon fails. 4. Passive replication may be used indeed fornon-deterministic processes. Que4.29. Explain data fragmentation with types. AKTU 2017- 18, Marks 10 Answer Fragmentation 1. It’s the corruption of a relation into fractions. 2. It permits to divide a single query into a set of multiplesub-queries that can execute resemblant on fractions. 3. Fragmentation is done according to the data selection patterns of operations running on the database. Fragmentation ways types are as follows 1. Vertical fragmentation a. It divides a relation into fractions which contain a subset of attributes of a relation along with the primary crucial trait of the relation. b. The purpose of perpendicular fragmentation is to partition a relation into a set of lower relations to enable stoner operations to run on only one scrap. 2. Vertical fragmentation a. It divides a relation into fractions along its tuples. Each scrap is a subset of tuples of a relation. b. It identifies some specific rows grounded on some criteria and marks it as a scrap. colorful vertical fragmentation ways are Primary vertical fragmentation This type of fragmentation is done where the tables in a database are neither joined nor have dependences . So, no relationship exists among the tables. ii. deduced vertical fragmentation deduced vertical fragmentation is used for parent relation. It’s used where tables are connected with the help of foreign keys. It ensures that the fractions which are joined together are put on the same point. 3. mongrel/ mixed fragmentation a. The mixed/ mongrel fragmentation is combination of vertical and perpendicular fragmentations. b. This type is most complex one, because both types are used in vertical and perpendicular fragmentation of the DB operation. c. The original relation is attained back by join or union operations. Que4.30. What are distributed database? List advantages and disadvantages of data replication and data fragmentation. Explain with a suitable illustration, what are the differences in replication and fragmentation translucency? OR Explain the types of distributed data storehouse. OR What are distributed database? List advantage and disadvantage of data replication and data fragmentation. Answer Distributed database ReferQ.4.26, runner 4 – 23A, Unit- 4. Advantages of data replication Vacuity If one of the spots containing relation r fails, also the relation r can be set up in another point. therefore, the system can continue to process queries involving ‘ r ’, despite the failure of one point. ii. Increased community Number of deals can read relation r in parallel. The further clones of ‘ r ’ there are, the lesser community is achieved. Disadvantages of data replication Increased outflow on update The system must insure that all clones of a relation r are harmonious; else, incorrect calculation may affect. therefore, whenever r is streamlined, the update must be propagated to all spots containing clones. The result is increased above. Advantages of data fragmentation Parallelized prosecution of queries by different spots is possible. ii. Data operation is easy as fractions are lower compare to the complete database. iii. Increased vacuity of data to the druggies queries that are original to the point in which the data stored. iv. As the data is available near to the place where it’s most constantly used, the effectiveness of the system in terms of query processing, sale processing is increased. Data that aren’t needed by original operations aren’t stored locally. It leads to reduced data transfer between spots, and increased security. Disadvantages of data fragmentation i. The performance of global operation that requires data from several fractions located at different spots may be slower. ii. Integrity control may be more delicate if data and functional dependences are fractured and located at different spots. Que4.31. bandy the types of distributed database. Answer Distributed databases are classified as 1. Homogeneous distributed database a. In this, all spots have identical database operation system software. b. All spots are apprehensive of one another, and agree to co-operate in processing stoner’s requests. 2. miscellaneous distributed database a. In this, different spots may use different schemas, and different database operation system software. b. The spots may not be apprehensive of one another, and they may give only limited installations for co-operation in sale processing. Que4.32. What’s concurrency control? Why it’s demanded in database system. OR Explain concurrency control. Why it’s demanded in database system? Answer 1. Concurrency Control( CC) is a process to insure that data is streamlined rightly and meetly when multiple deals are coincidently executed in DBMS. 2. It’s a medium for correctness when two or further database deals that pierce the same data or dataset are executed coincidently with time imbrication. 3. In general, concurrency control is an essential part of sale operation. Concurrency control is demanded 1. To insure thickness in the database. 2. To help following problem Lost update i. A alternate sale writes a alternate value of a data item on top of a first value written by a first concurrent sale, and the first value is lost to other deals running coincidently which need, by their priority, to read the first value. ii. The deals that have read the wrong value end with incorrect results. Dirty read Deals read a value written by a sale that has been latterly abandoned. ii. This value disappears from the database upon repeal, and should not have been read by any sale( “ dirty read ”). iii. The reading deals end with incorrect results.
