Research of Efficiency of Multi-core Computers with Shared Memory

Nedzelskyi, D. O.; Derkach, Maryna; Tatarchenko, Yehor; Safonova, Svitlana; Shumova, Larisa; Kardashuk, Volodymyr

doi:10.1109/ficloudw.2019.00032

Cited by 1 publication

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, without knowing the exact structure of the program being executed (sequential ordering of arbitrary commands, which commands depend on the results of previous commands, how many conditional transitions, how many commands of each kind, etc. ), it is impossible to predict the degree of parallelism of the execution of commands in functional devices and the combination them with accesses to the memory subsystem [14]. The entire execution time of the program will be no less than the execution time of the longest stage of the parallel cycle.…”

Section: The Model Of Program Execution In One Corementioning

confidence: 99%

A Load Factor and its Impact on the Performance of a Multicore System with Shared Memory

Nedzelky

Derkach

Skarga-Bandurova

et al. 2021

2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applicat

View full text Add to dashboard Cite

The paper investigates the influence of the load factor of the shared memory on the efficiency of multicore systems. Typically, all cores serve threads of one program in parallel by the OpenMP programming technology or execute independent programs. There are no interactions between threads and independent programs, but conflicts can occur when accessing the shared memory. Models of program execution in one core and a multicore computer are developed, considering the probabilities of successful calls and service times of all levels of the shared memory subsystem. The load factor of the first level cache is determined through the ratio of the L1 cache load time to the total execution time of the program. The execution of various types of programs is simulated. A technique for the acceleration coefficient of a multicore computer based on the total load factor of the shared memory has been proposed. Based on this insight, we apply our model to determine the acceleration coefficient for 4-, 8-, 12-and 16core systems for different combinations of system parameters.

show abstract

Section: The Model Of Program Execution In One Corementioning

confidence: 99%