| A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999. |
....models vary in their ability to account for self and cross interference misses. In this study, we see that on current and future machines, whose cache sizes continue to grow, conflict misses do not contribute significantly to accurate miss modeling. Work in sparse compilers, e.g. Bik et al. [7] and the Bernoulli compiler [37] complements our own work. These projects focus on the expression of sparse kernels and data structures for code generation, and will likely prove valuable to generating our implementations. One distinction of our work is our use of a hybrid o# line, on line model ....
A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
....that come from a variety of domains. Furthermore, we explicitly model execution time (instead of just modeling misses) in order to evaluate the extent to which our tuned implementations achieve optimal performance. Finally, we mention examples of work in the sparse compiler literature by Bik [2], Pugh and Shpeisman [25] and the Bernoulli compiler [28] The first analyzes matrices for high level structure using techniques complementary those that we consider; the latter two consider sparse code specification and generation issues, but do not specialize for specific matrix structures. The ....
A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
....formats that we are aware of, and it also permits users to define new formats. Our code synthesis algorithm is able to exploit the index structure of sparse matrix formats and generate code competitive with hand written library codes for the BLAS routines. Automatic selection of sparse formats [3] for particular applications is an interesting extension to the work described here. One possibility is to make the compiler responsible for making this selection using cost estimation rules like the ones described in Section 4. Another possibility is to use an empirical optimization approach ....
Aart Bik and Harry Wijshoff. Automatic nonzero structure analysis. SIAM Journal of Computing, 28(5):1576--1587, 1999.
No context found.
A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
No context found.
Aart Bik and Harry Wijshoff. Automatic nonzero structure analysis. SIAM Journal of Computing, 28(5):1576--1587, 1999.
No context found.
A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
No context found.
A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal of Computing, 28(5):1576--1587, 1999.
No context found.
Aart J. C. Bik and Harry A. G. Wijshoff. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
No context found.
Aart J. C. Bik and Harry A. G. Wijshoff. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
No context found.
Aart J. C. Bik and Harry A. G. Wijshoff. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
No context found.
A. J. C. Bik and H. A. G. Wijsho. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576-1587, 1999.
No context found.
A. J. C. Bik and H. A. G. Wijsho#. Automatic nonzero structure analysis. SIAM Journal on Computing, 28(5):1576--1587, 1999.
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