Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems
Main Article Content
Abstract
Recently, reconfigurable systems based on field programmable logic devices (FPLDs) have been widely used in high-performance computing. The paper discusses issues related to the experimental research of a shared memory subsystem with a limited queue length of specialized reconfigurable multiprocessor systems using the developed mathematical modelling method. The paper presents the results of the method proposed by the authors for modelling multiprocessor systems based on open queuing networks with limited queue lengths. Based on these conditions, as well as the architectural features of the investigated processor-memory subsystem, expressions are calculated to estimate the exchange time and the resulting delays at each exchange stage. During the research, the main attention was paid to the dependence of the increase in the number of processor nodes in the processor-memory subsystem. As a result, the data obtained showed that the processor growth significantly affects the exchange time, creating a significant load on the common bus, as well as increasing delays at the stages when request transfer operation from the processor to the memory is performed. At the same time, the inadequate behaviour of experimental results and inaccuracy of their values when using the basic modelling method are explicitly tracked, which is reflected in the obtained graphs. Computational experiments were carried out to calculate the probabilistic-temporal characteristics of the "processor-memory" subsystem using the developed mathematical modelling methods. Based on the experimental results, it was determined that the delays occurring in subsystem's nodes and the time of exchange between the processor and memory modules depend on the query parameters and the processor-memory subsystem’s architectural characteristics.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
1. Proposal of Policy for Free Access Periodics
Authors whom publish in this magazine should agree to the following terms:
a. Authors should keep the copyrights and grant to the magazine the right of the first publication, with the work simultaneously permitted under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 that allows the sharing of the work with recognition of the authorship of the work and initial publication in this magazine.
b. Authors should have authorization for assuming additional contracts separately, for non-exclusive distribution of the version of the work published in this magazine (e.g.: to publish in an institutional repository or as book chapter), with recognition of authorship and initial publication in this magazine.
c. Authors should have permission and should be stimulated to publish and to distribute its work online (e.g.: in institutional repositories or its personal page) to any point before or during the publishing process, since this can generate productive alterations, as well as increasing the impact and the citation of the published work (See The Effect of Free Access).
Proposal of Policy for Periodic that offer Postponed Free Access
Authors whom publish in this magazine should agree to the following terms:
a. Authors should keep the copyrights and grant to the magazine the right of the first publication, with the work simultaneously permitted under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 [SPECIFY TIME HERE] after the publication, allowing the sharing of the work with recognition of the authorship of the work and initial publication in this magazine.
b. Authors should have authorization for assuming additional contracts separately, for non-exclusive distribution of the version of the work published in this magazine (e.g.: to publish in institutional repository or as book chapter), with recognition of authorship and initial publication in this magazine.
c. Authors should have permission and should be stimulated to publish and to distribute its work online (e.g.: in institutional repositories or its personal page) to any point before or during the publishing process, since this can generate productive alterations, as well as increasing the impact and the citation of the published work (See The Effect of Free Access).
d. They allow some kind of open dissemination. Authors can disseminate their articles in open access, but with specific conditions imposed by the editor that are related to:
Version of the article that can be deposited in the repository:
Pre-print: before being reviewed by pairs.
Post-print: once reviewed by pairs, which can be:
The version of the author that has been accepted for publication.
The editor's version, that is, the article published in the magazine.
At which point the article can be made accessible in an open manner: before it is published in the magazine, immediately afterwards or if a period of seizure is required, which can range from six months to several years.
Where to leave open: on the author's personal web page, only departmental websites, the repository of the institution, the file of the research funding agency, among others.
References
Ali, H., Tariq, U. U., Hardy, J., Zhai, X., Lu, L., Zheng, Y., ... & Antonopoulos, N. (2021). A survey on system level energy optimisation for MPSoCs in IoT and consumer electronics. Computer Science Review, 41, 100416.
Cudennec, L., & Trabelsi, K. (2020, August). Experiments Using a Software-Distributed Shared Memory, MPI and 0MQ over Heterogeneous Computing Resources. In European Conference on Parallel Processing (237-248). Springer, Cham.
Ghose, S., Hsieh, K., Boroumand, A., Ausavarungnirun, R., & Mutlu, O. (2019). The processing-in-memory paradigm: Mechanisms to enable adoption. In Beyond-CMOS Technologies for Next Generation Computer Design (133-194). Springer, Cham.
Lakhani, K. J. (2020). Using GPUDirect RDMA and IPC Shared Memory for Direct DMA in a Client-server (FPGA-GPU) System: One Step Closer to a Fast and Robust DNA Sequencer (Doctoral dissertation, University of California, Davis).
Lant, J., Concatto, C., Attwood, A., Pascual, J. A., Ashworth, M., Navaridas, J., ... & Goodacre, J. (2019). Enabling shared memory communication in networks of MPSoCs. Concurrency and Computation: Practice and Experience, 31(21), e4774.
Lopes, A. S., Brandalero, M., Beck, A. C., & Pereira, M. M. (2019, November). Generating optimized multicore accelerator architectures. In 2019 IX Brazilian Symposium on Computing Systems Engineering (SBESC) (1-8). IEEE.
Martyshkin, A. I. (2021). Pilot Model of the Embedded Reconfigurable Real Time Computing System. International Journal of Engineering Research and Technology, 13(12), 4635-4645.
Martyshkin, A. I., Pashchenko, D. V., Trokoz, D. A., Sinev, M. P., & Svistunov, B. L. (2020). Using queuing theory to describe adaptive mathematical models of computing systems with resource virtualization and its verification using a virtual server with a configuration similar to the configuration of a given model. Bulletin of Electrical Engineering and Informatics, 9(3), 1106-1120.
Martyshkin, A. (2019). Software package for determining characteristics of task managers of reconfigurable computer systems using priority queueing networks. Revista Inclusiones, 463-474.
Nguyen, Q. M., & Sanchez, D. (2021, October). Fifer: Practical Acceleration of Irregular Applications on Reconfigurable Architectures. In MICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture (1064-1077).
Reis, J. G., & Fröhlich, A. A. (2020). Towards deterministic FPGA reconfiguration. International Journal of Embedded Systems, 13(2), 236-253.
Sinha, M., Harsha, G. S., Bhattacharyya, P., & Deb, S. (2021). Design space optimization of shared memory architecture in accelerator-rich systems. ACM Transactions on Design Automation of Electronic Systems (TODAES), 26(4), 1-31.