Abstract
Pointer-intensive and sparse numerical computations typically display irregular memory access behavior. This work presents a mathematical model, called the Self-tuning Adaptive Predictor (SAP), to characterize the behavior of load instructions in procedures with pointer-based data structures by using procedure call boundaries as the fundamental sampling frequency. This model incorporates information about the history of specific load instructions (temporal locality) and their neighboring loads (spatial locality) using a least-squares minimization approach. Simulation results on twelve of the most time-consuming procedures with pointer-based data structures from five of the SPEC2000 integer benchmark programs show that these pointer-based data structures demonstrate surprisingly regular memory access patterns. The prediction error at steady-state is within [-6%, +6%] on average.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Chilimbi, T.M., Larus, J.R.: Using generational garbage collection to implement cache-conscious data placement. In: Proceedings of the First International Symposium on Memory Management. ACM SIGPLAN Notices, vol. 34(3) (October 1998)
Ding, C., Kennedy, K.: Improving cache performance in dynamic applications through data and computation reorganization at run time. In: Proceedings of the SIGPLAN 1999 Conference on Programming Language Design and Implementation, Atlanta, GA (May 1999)
Zhang, H., Martonosi, M.: A Mathematical Cache Miss Analysis for Pointer Data Structures. In: SIAM Conference on Parallel Processing for Scientific Computing (March 2001)
Draper, N.R., Smith, H.: Applied Regression Analysis, 2nd edn. John Wiley & Sons, Chichester (1981)
Burger, D.C., Austin, T.M., Bennett, S.: Evaluating future Microprocessors: The SimpleScalar Tool Set. Technical Report CS-TR-96-1308, University of Wisconsin-Madison (July 1996)
Wu, K., Sendag, R., Lilja, D.J.: Using a Self-tuning Adaptive Predictor to Characterize the Regularity of Memory Accesses in Pointer-Intensive Application Programs. University of Minnesota Technical Report: ARCTiC 03-04 (April 2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wu, K., Sendag, R., Lilja, D.J. (2003). Exploring Memory Access Regularity in Pointer-Intensive Application Programs. In: Liu, J., Cheung, Ym., Yin, H. (eds) Intelligent Data Engineering and Automated Learning. IDEAL 2003. Lecture Notes in Computer Science, vol 2690. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-45080-1_65
Download citation
DOI: https://doi.org/10.1007/978-3-540-45080-1_65
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-40550-4
Online ISBN: 978-3-540-45080-1
eBook Packages: Springer Book Archive