An Example of Correct Global Trace Generation

The memory reference trace of a parallel program is strongly dependent on the memory architecture and memory management policy environment. A change in that environment can change the addresses in the trace as well as the interleaving of the addresses. We have examined the issue of ensuring that the generated global trace is correct for the new environment. In this paper we summarize the key results and informally illustrate the proposed method using an example parallel program.

[1]  Alfred V. Aho,et al.  Compilers: Principles, Techniques, and Tools , 1986, Addison-Wesley series in computer science / World student series edition.

[2]  Leslie Lamport,et al.  How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs , 2016, IEEE Transactions on Computers.

[3]  Ned Snell Fiserv Inc. , 1993 .

[4]  Carla Schlatter Ellis,et al.  Accuracy of Memory Reference Traces of Parallel Computations in Trace-Driven Simulation , 1992, IEEE Trans. Parallel Distributed Syst..

[5]  Alan Jay Smith Cache Evaluation and the Impact of Workload Choice , 1985, ISCA.

[6]  Mark A. Holliday Techniques for Cache and Memory Simulation Using Address Reference Traces , 1991, Int. J. Comput. Simul..