| R. A. Uhlig and T. N. Mudge, "Trace-Driven Memory Simulation: A Survey," ACM Computing surveys, vol. 29, no. 2, 1997, pp. 128-170. |
....I am and everything I do possible. Chapter 1 Introduction A current trend in computer architecture is that the speed increase of processors continually outpaces that of main memory. Since 1985, the rate of improvement in processor speed has been 40 to 100 per year [Hennessy 96, Patterson 97, Uhlig 97] Main memory which is composed of commodity dynamic random access memory (DRAM) devices, has a slower rate of improvement of only about 7 to 11 per year [Hennessy 96, Patterson 97, Uhlig 97] Thus, while the absolute memory latency is actually decreasing, the relative memory latency from the ....
.... 1985, the rate of improvement in processor speed has been 40 to 100 per year [Hennessy 96, Patterson 97, Uhlig 97] Main memory which is composed of commodity dynamic random access memory (DRAM) devices, has a slower rate of improvement of only about 7 to 11 per year [Hennessy 96, Patterson 97, Uhlig 97] Thus, while the absolute memory latency is actually decreasing, the relative memory latency from the point of view of the processor is increasing each year. Because applications need to access main memory, the increasing memory latency will lead to the circumstance, at the extreme, where ....
[Article contains additional citation context not shown here]
R. Uhlig and T. Mudge. Trace-Driven Memory Simulation: A Survey. ACM Computing Surveys, 29(2):128--170, June 1997.
....is executed. With trace driven simulation the code is instrumented to obtain a file which contains all the references to memory generated by its execution. Then, the file is fed to the simulator which provides statistics related to the number and characteristics of the movement of data generated [21]. The main benefit of this approach is the possibility of a parametric study of the memory by varying its characteristics. The disadvantage is the large amount of time and memory required. The second possibility consists on hardware monitoring tools. Some processors include hardware elements that ....
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29(2):128--170, June 1997.
....appropriate aspects of program behavior to investigate. Most prior work has focused on time independent aspects of program activity. In these approaches, event ordering and interleaving are of prime importance. An example of such work is the trace based analysis adopted by most researchers [74, 78, 29]. In trace based analysis, program memory behavior is typically represented as collections of memory address traces. Memory addresses and their access ordering are captured in these traces, but the time intervals between accesses are missing, therefore can not be exploited. In contrast to ....
R. A. Uhlig and T. N. Mudge. Trace-Driven Memory Simulation: A Survey. ACM Computing Surveys, 29(2):128--170, 1997.
....know the cache behavior of the program region it would apply them to. In this paper, a method is devised which calculates the cache behavior of sequences of loop regions in the program. 1. 2 Cache Bottleneck Identi cation The most traditional way to measure a programs cache behavior is pro ling[14]. The program is instrumented so that during execution, each access to the memory is traced and fed into a cache simulator. This method gives exact results, but is very slow, since each individual memory access needs to be processed. Furthermore, the cache statistics are only measured for a ....
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: a survey. ACM Computing Surveys (CSUR), 29(2):128-170, 1997.
....8 Related Work Programs must exhibit sucient locality to achieve good cache performance. Compiler optimisations for improving the cache behaviour need to have detailed knowledge about the number and causes of cache misses. Such an information can be obtained by time consuming cache simulation [25] and architecturedependent hardware counters [1] Analytical methods use mathematical formulas to provide a characterisation of a program s cache behaviour so that we can not only obtain the number of cache misses but also reason about the causes of such misses from these formulas. The ultimate ....
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: a survey. ACM Computing Surveys, 29(3):128-170, Sept. 1997.
....for sensitivity analysis, are derived from example parameters from commercial processors and are described in the following section. The chapter concludes with a discussion of each of the parameters considered for the sensitivity analysis. 50 3. 2 Simulation environment A trace driven approach [48] was taken in this study. This involved the generation of address traces for the workloads and then a simulation of the traces on different cache and TLB configurations. 0x000080e8 0x000080ec 0xbfffc000 0x00080f0 0x0008108 d D . i i i Figure 3.1: A fragment of an ....
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29(2):128--170, 1997.
No context found.
R. A. Uhlig and T. N. Mudge, "Trace-Driven Memory Simulation: A Survey," ACM Computing surveys, vol. 29, no. 2, 1997, pp. 128-170.
No context found.
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: a survey. ACM Computing Surveys (CSUR), 29(2):128--170, 1997. 3.4
No context found.
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29(2):128--170, 1997.
No context found.
R.A. Uhlig and T.N. Mudge, "Trace-Driven Memory Simulation: A Survey", ACM Computing
No context found.
R. Uhlig and T. Mudge, "Trace-Driven Memory Simulation: A Survey, " ACM Computing Surveys, vol. 29, pp. 128--170, June 1997.
No context found.
R. Uhlig, T. Mudge, Trace-driven memory simulation: A survey, ACM Computing Surveys 29 (2) (1997) 128--170.
No context found.
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29(2):128--170, June 1997.
No context found.
R. Uhlig, T. Mudge, Trace-driven memory simulation: A survey, ACM Computing Surveys 29 (2) (1997) 128--170.
No context found.
Uhlig, R., Mudge, T.: Trace-Driven Memory Simulation: A Survey. ACM Computing Surveys 29 (1997) 128--170
No context found.
R. Uhlig and T. Mudge. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29:128--170, June 1997.
No context found.
R.A. Uhlig, T.N. Mudge. "Trace-Driven Memory Simulation: A Survey", ACM Computing Surveys, v.29, n.2, June 1997, pp. 128-170.
No context found.
R. Uhlig and T. Mudge. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29(2):128--170, June 1997.
No context found.
Richard A. Uhlig and Trevor N. Mudge. Trace-driven memory simulation: a survey. ACM Computing Surveys (CSUR), 29(2):128--170, 1997.
No context found.
R. A. Uhlig and T. N. Mudge. Trace-driven Memory Simulation: a Survey. ACM Computing Surveys, 29(2):128 -- 170, June 1997. 16
No context found.
R.A. Uhlig and T.N. Mudge, "Trace-Driven Memory Simulation: A Survey," ACM Computing Surveys, 29(2), pp. 128-170, June 1997.
No context found.
R. UHLIG and T. MUDGE. Trace-driven memory simulation: A survey. ACM Computing Surveys, 29(2):128--170, Jun 1997.
No context found.
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: a survey. ACM Computing Surveys (CSUR), 29(2):128--170, 1997. 3.4
No context found.
Uhlig et al., "Trace-driven memory simulation: a survey", ACM Computing Surveys, Vol.29, No. 2, June 1997.
No context found.
R. A. Uhlig and T. N. Mudge. Trace-driven memory simulation: a survey. ACM Computing Surveys, 29(3):128--170, Sept. 1997.
First 50 documents Next 50
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