J. Haskins and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proc. of the 2001.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....computer architecture research typically requires all of the statistics generated by a simulator such as sim outorder. With this need in mind, using a less detailed simulator often is not an option. Various techniques have been proposed which either take represen tative samples of the execution [6] or which combine statistical and functional simulation [3] 12] 13] In our case, we want full detail execution from start to finish of the benchmark program. That means our best option is to find a quantitatively defensible way to reduce the input datasets, and, consequently, the runtimes, ....

....The reference dataset is intended to give a complete evaluation of the host computer system s performance. These datasets are similar to our desired new workload except on a much larger time scale. We began our analysis by compiling the benchmarks with the SimpleScalar version of gcc (version 2. 6.3) on a Sun UltraSparc running Solaris. This modified version of gcc builds binaries for the simulator s PISA architecture. We compiled at four different optimization levels, O0 through 03. We then ran each SPEC dataset (test, train, and re f) with sire fast, a simple simulator for determining ....

[Article contains additional citation context not shown here]

John Haskins and Kevin Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Interna- tional Conference on Computer Design (ICCD), September 2001.

....as cycle accurate simulation, modeling all cache and branch predictor interactions is still costly. One viable method for further accelerating sampled simulations is to avoid full warm up by only modeling those interactions that occur within a given number of instructions prior to the sample [2, 3, 5]. Our technique makes the determination of when to engage cache and branch predictor warm up by exploiting memory reference reuse latencies (MRRL) a measurement of the number of instructions that elapse between successive references to the same address. We have developed software that ....

....that are known to contain good state; using stale state from the previous sample; and flushing state but estimating how much error this introduces. The warm up acceleration methods proposed by [2, 5] however, may compromise the accuracy of the pre sample state initialization. Haskins et al. [3] propose a warm up acceleration technique called Minimal Subset Evaluation (MSE) which exploits the observation that only the most recent memory references prior to a sample are germane to memory references that occur during the sample itself. The MSE technique uses formulas derived from ....

[Article contains additional citation context not shown here]

J. W. Haskins, Jr. and K. Skadron. Minimal Subset Evaluation: Rapid Warm-up for Simulated Hardware State. In Proceedings of the International Conference on Computer Design, Sept. 2001.

....Conte, Hirsch, Menezes, and Hwu [4, 5] describe various techniques for reducing cold start bias for cache and branchpredictor simulation. Unfortunately, these prior techniques for dealing with cold start bias are heuristics whose accuracy can only be verified experimentally. Haskins and Skadron [13] describe Minimal Subset Evaluation (MSE) a two pass method that, for a user specified probability of accuracy, probabilistically determines a minimally By sampling error, we refer to the deviation between quantitative results obtained by full execution relative to results obtained by executing ....

J. W. Haskins, Jr. and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proceedings of the 2001.

No context found.

J. Haskins and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proc. of the 2001.

No context found.

J. Haskins and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proc. of the 2001.

No context found.

J. W. Haskins and K. Skadron, "Minimal Subset Evaluation: Rapid Warm-Up for Simulated Hardware State," In Proceedings of the International Conference on Computer Design, September 2001.

No context found.

J. W. Haskins and K. Skadron, "Minimal Subset Evaluation: Rapid Warm-Up for Simulated Hardware State," In Proceedings of the International Conference on Computer Design, September 2001.

No context found.

J. Haskins and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proceedings of the 2001.

No context found.

J. Haskins Jr. and K. Skadron. Minimal Subset Evaluation: Rapid Warm-up for Simulated Hardware State. In International Conference on Computer Design, pages 32--39, Austin, Texas, September 2001.

No context found.

J. Haskins and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proceedings of the 2001.

No context found.

J. Haskins and K. Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In Proceedings of the 2001.

No context found.

John Haskins and Kevin Skadron. Minimal subset evaluation: Rapid warm-up for simulated hardware state. In International Conference on Computer Design (ICCD), September 2001.

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