Ahamad, M. et al. (1991) Causal memory. In Proc. 5th Int. Workshop on Distributed Algorithms, pp. 9--30.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....such as the one shown above. This model generalizes the strict memory model potentially needing a lower number of messages to operate, and increasing the parallelism of the system. The sufficient memory model is similar to the sequentially consistent memory such as defined by M. Ahamad et al. [1]. Also, the sufficient memory model is similar to the strict model in the sense that both allow only executions having a consistent observation on which all processes agree. As a matter of fact the 5 P2 P1 P2 P2 Computation considering global time First view of the computation Second ....

Ahamad, M., Burns, J.E., Hutto, P.H., Neiger, G. Causal Memory. Proceedings of the 5th International Workshop on Distributed Algorithms, Oct. 1991.

....system states. The purpose of a data consistency protocol is to provide a consistent view of the shared memory in the presence of multiple copies. Data consistency protocols for implementing shared memories in message passing systems have received much attention [AHJ91, ABND90, AF92, HA90, LH89, ABHN91, MRZ95, RM93] Data consistency protocols implementing strong consistency require that all copies of an object be identical at all times. This, however, may be a very stringent consistency requirement. Lamport proposed the notion of sequential consistency, which states that a shared memory ....

M. Ahamad, J. Burns, P. Hutto, and G. Neiger. Causal memory. In Proceedings of 5th WIDAG, LNCS 579, pages 9--30, 1991.

....views of a given computation. Definition 11 We will define the sufficient memory model MSU(P;M;M 0 ) as the one for which all the well formed behaviors admit a sufficient permutation. The sufficient memory model is similar to the sequentially consistent memory such as defined by M. Ahamad et al. [2]. Also, the sufficient memory model is similar to the strict model in the sense that both allow only executions having a consistent observation on which all processes agree. As a matter of fact the sufficient memory model is the most general one for which this is true. This is a consequence of ....

Ahamad, M., Burns, J.E., Hutto, P.H., Neiger, G. Causal Memory. Proceedings of the 5th International Workshop on Distributed Algorithms, Oct. 1991.

....We call it reads from relation and sometimes explicitly denote this relation by w i (x) rf r j (x) Note that irreflexive and transitive relation implies antisymmetric relation. Thus, this relation is equivalent to an irreflexive directed acyclic graph (or an irreflexive partially ordered set) [3]. An execution history (or history) is a prefix of a complete history. Throughout the paper, for relation with transitivity , we use op i op j to denote relation either op i = op j or op i op j . A history H= H; H ) is a sequential history if H is a total order. A history ....

Ahamad, M., Burns, J., Hutto, P., and Neiger, G. Causal memory. In Proceedings of 5th IWDAG, LNCS 579, pages 9--30, 1991.

....a read allows to obtain a value of the object. A write of value v into object x by process P i is denoted w i (x)v; similarly a read of x by process P j is denoted r j (x)v where v is the value returned by the read operation; op will denote either r (read) or w (write) For simplicity, as in [26, 4, 32], we assume all values written 1 Moreover, it is worth noting that this set of formal definitions is based on very few (and simple) mathematical notions, namely: partial order, linear extension, suborder and legality (of read operations) Irisa Consistency Criteria for Shared Memories 5 into ....

....(2) realizes writer writer mutual exclusion (QW1 x and QW2 x are write quorums on x) 8x 2 X : 8 (QR x ; QW x ) QR x QW x 6= 1) 8x 2 X : 8 (QW1 x ; QW2 x ) QW1 x QW2 x 6= 2) 5 Causal Consistency 5. 1 Definition Causal consistency has first been introduced by Ahamad et al. in 1991 [4], and then studied by several authors [5, 3, 32] It defines a consistency criterion strictly weaker than sequential consistency, and allows for a wait free implementation of read and PI n968 10 M. Raynal and A. Schiper write operations in a distributed environment (i.e. causal consistency ....

M. Ahamad, J.E. Burns, P.W. Hutto, and G. Neiger. Causal memory. in Proc. 5th Int. Workshop on Distributed Algorithms (WDAG-5), pages 9--30. Springer Verlag, LNCS 579, 1991.

....a read allows to obtain a value of the object. A write of value v into object x by process P i is denoted w i (x)v; similarly a read of x by process P j is denoted r j (x)v where v is the value returned by the read operation; op will denote either r (read) or w (write) For simplicity, as in [13, 3, 19], we assume all values written into an object x are distinct. Moreover, the parameters of an operation are omitted when they are not important. Each object has an initial value; it is assumed that this value has been assigned by an initial fictitious write operation. 1 Moreover, it is worth ....

....page, and no one else has a copy of the page. These mechanisms ensure atomicity (i.e. mutual exclusion) between any couple of read and write operations, and any couple of write operations. 5 Causal Consistency 5. 1 Definition Causal consistency has first been introduced by Ahamad et al. in 1991 [3], and then studied by several authors [4, 19] It defines a consistency criterion strictly weaker than sequential consistency, and allows for a wait free implementation of read and write operations in a distributed environment (i.e. causal consistency allows for cheap read write operations) With ....

M. Ahamad, J.E. Burns, P.W. Hutto, and G. Neiger. Causal memory. in Proc. 5th Int. Workshop on Distributed Algorithms (WDAG-5), pages 9--30. Springer Verlag, LNCS 579, 1991.

....x p : m; h8q : q 6= p : X q : X q [ fmgi. 3. Process M p;x is as follows: repeat if X p 6= fg then x p ; X p : m; X p Gamma fmg where m 2 X p forever 3 Causal Memory This kind of non atomic memory was defined by Ahamad, Hutto, and John [2] and later by Ahamad, Burns, Hutto, and Neiger [1]. A causality relation is maintained across read and write events. There are two components of this causality relation: the first is based on program order (this also occurs in the pipelined RAM model) and the second is based on the values read by the processes. In order to model this memory, we ....

M. Ahamad, J. E. Burns, P. W. Hutto, and G. Neiger. Causal memory. In Proceedings of the Fifth International Workshop on Distributed Algorithms, Lecture Notes in Computer Science. Springer-Verlag, Berlin, October 1991.

....to broadcast messages. Note that cbc can always do a local delivery of a message immediately. Hence, both reads and writes are executed immediately. Also, note that by using the cbc service of vscl, the code of the algorithm becomes very simple since, unlike other implementations of causal memory [4, 5, 6], we do not need to implement causal broadcast as part of our memory consistency protocol. The proof of this algorithm is very similar to the proof of the algorithm for sequential consistency, and is therefore dropped from this paper. 5 Discussion In recent years it becomes evident that ....

M. Ahamad, J. Burns, P. Hutto, and G. Neiger. Causal Memory. In 5th International Workshop on Distributed Algorithms, Greece, October 1991.

.... claimed that a memory consistency criterion is a contract between the memory system and application programs [11] Three main consistency criteria have been proposed in the literature: atomic consistency [9] also called linearizability [7] sequential consistency [8] and causal consistency [3]. In all three cases a read operation returns the last value assigned to the variable (or written into the object) The three consistency criteria differ however in the definition of the last write operation. Atomic consistency is the more restrictive of the three consistency criteria: it requires ....

....to a sequential execution 5 . More formally, we define sequential consistency in the following way. Definition. Sequential consistency. A history b H = H; H ) is sequentially consistent if it admits a linear extension 6 b S in which all reads are legal. 4 Other definitions of legality [3, 4] eliminate only the possibility of an intervening write (w(x)u) between the writing of some value (w(x)v) and a reading of the same value (r(x)v) 5 In his definition, Lamport assumes that the process order relation defined by the program (see point (2) of the definition) is maintained in the ....

[Article contains additional citation context not shown here]

M. Ahamad, J.E. Burns, P.W. Hutto, and G. Neiger. Causal Memory. In Proc. 5th Intl. Workshop on Distributed Algorithms (WDAG-5), pages 9--30. Springer Verlag, LNCS 579, 1991.

.... claimed that a memory consistency criterion is a contract between the memory system and application programs [21] Three main consistency criteria have been proposed in the literature: atomic consistency [19] also called linearizability [14] sequential consistency [17] and causal consistency [3]. In all three cases a read operation returns the last value assigned to the variable (or written into the object) The three consistency criteria differ however in the definition of the last write operation. Atomic consistency is the more restrictive of the three consistency criteria: it requires ....

....in the order specified by its program . This informal definition states that the execution of a program is sequentially consistent if it is equivalent to a sequential execution 4 . More formally, we define sequential consistency in the following way. 3 The usual definition of legality [3, 4] eliminates only the possibility of an intervening write (w(x)u) between the writing of some value (w(x)v) and a reading of the same value (r(x)v) Our definition of legality allows for a simpler definition of causal consistency (see Section 2.4) 4 In his definition, Lamport assumes that the ....

[Article contains additional citation context not shown here]

M. Ahamad, J.E. Burns, P.W. Hutto, and G. Neiger. Causal Memory. In Proc. 5th Intl. Workshop on Distributed Algorithms (WDAG-5), pages 9--30. Springer Verlag, LNCS 579, 1991.

....in every process. P j represented by the vector M j [ j] be greater than the vector timestamp M [ j] of a message that is accepted by P j . Note that if only messages directed towards a single location must be causally ordered (e.g. to realize a causally consistent monitor or causal memory [1]) or if causally ordered broadcasts are considered [8, 37] then the matrix can be reduced to a vector by simply ignoring its second dimension. Interestingly, causal order message delivery can also be implemented without vectors or matrices by using input output message buffers that communicate ....

Ahamad M, Burns JE, Hutto PW, Neiger G: Causal Memory. In: P Spirakis, S Toueg (eds) Proc 5th International Workshop on Distributed Algorithms. Springer, Berlin Heidelberg New York, LNCS 579: 9--30 (1991)

....We call it readsfrom relation and sometimes explicitly denote this relation by w i (x) rf r j (x) Note that irreflexive and transitive relation implies antisymmetric relation. Thus, this relation is equivalent to an irreflexive directed acyclic graph (or an irreflexive partially ordered set) [3]. An execution history (or history) is a prefix of a complete history. Throughout the paper, for a relation with transitivity , we use op i op j to denote relation either op i = op j or op i op j . ffl A history H= H; H ) is a sequential history if H is a total order. ffl A ....

Ahamad, M., Burns, J., Hutto, P., and Neiger, G. Causal memory. In Proceedings of 5th IWDAG, LNCS 579, pages 9--30, 1991.

No context found.

Ahamad, M. et al. (1991) Causal memory. In Proc. 5th Int. Workshop on Distributed Algorithms, pp. 9--30.

No context found.

M. Ahamad. J.E. Burns, P.W. Hutto, and G. Neiger. Causal memory. In Proc. of the 11th Int'l Conf. on Distributed Computing Systems, pp. 274--281, May 1991.

No context found.

M. Ahamad, J. Burns, P. Hutto, and G. Neiger. Causal Memory. In 5th International Workshop on Distributed Algorithms, Greece, October 1991.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC