z-logo
open-access-imgOpen Access
Specifying memory consistency of write buffer multiprocessors
Author(s) -
Lisa Highám,
LillAnne Jackson,
Jalal Kawash
Publication year - 2007
Publication title -
acm transactions on computer systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.343
H-Index - 70
eISSN - 1557-7333
pISSN - 0734-2071
DOI - 10.1145/1189736.1189737
Subject(s) - computer science , consistency model , cache coherence , sequential consistency , write buffer , parallel computing , memory model , consistency (knowledge bases) , scalability , shared memory , out of order execution , multiprocessing , weak consistency , equivalence (formal languages) , causal consistency , abstraction , programming language , strong consistency , cpu cache , correctness , operating system , cache , philosophy , mathematics , artificial intelligence , estimator , cache algorithms , linguistics , epistemology , statistics
Write buffering is one of many successful mechanisms that improves the performance and scalability of multiprocessors. However, it leads to more complex memory system behavior, which cannot be described using intuitive consistency models, such as Sequential Consistency. It is crucial to provide programmers with a specification of the exact behavior of such complex memories. This article presents a uniform framework for describing systems at different levels of abstraction and proving their equivalence. The framework is used to derive and prove correct simple specifications in terms of program-level instructions of the sparc total store order and partial store order memories.The framework is also used to examine the sparc relaxed memory order. We show that it is not a memory consistency model that corresponds to any implementation on a multiprocessor that uses write-buffers, even though we suspect that the sparc version 9 specification of relaxed memory order was intended to capture a general write-buffer architecture. The same technique is used to show that Coherence does not correspond to a write-buffer architecture. A corollary, which follows from the relationship between Coherence and Alpha, is that any implementation of Alpha consistency using write-buffers cannot produce all possible Alpha computations. That is, there are some computations that satisfy the Alpha specification but cannot occur in the given write-buffer implementation.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here
Accelerating Research

Address

John Eccles House
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom