Optimizing Big Data Network Transfers in FPGA SoC Clusters: TECBrain Case Study

Luis G. León-Vega, Kaleb Alfaro-Badilla, Alfonso Chacón-Rodríguez, Carlos Salazar-García

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Spiking Neural Network (SSN) simulators based on clusters of FPGA-based System-on-Chip (SoC) involve the transmission of large amounts of data (from hundreds of MB to tens of GB per second) from and to a data host, usually a PC or a server. TECBrain is an SNN simulator which currently uses Ethernet for transmitting results from its simulations, which can potentially take hours if the effective connection speed is around 100 Mbps. This paper proposes data transfer techniques that optimize data transmissions by grouping data into packages making the most of the payload size and the use of thread-level parallelism, trying to minimize the impact of multiple clients transmitting at the same time. The proposed method achieves its highest throughput when inserting simulation results directly into a No-SQL database. Using the proposed optimization techniques over an Ethernet connection, the minimum overhead reached is 2.93% (out of the theoretical 2.47%) for five nodes sending data simultaneously from C++, with speeds up to 95 Mbps on a network at 100 Mbps. Besides, the maximum database insertion speed reached is 32.5 MB/s, using large packages and parallelism, which is 26% of the bandwidth of the connection link at 1 Gbps.

Original languageEnglish
Title of host publicationHigh Performance Computing - 6th Latin American Conference, CARLA 2019, Revised Selected Papers
EditorsJuan Luis Crespo-Mariño, Esteban Meneses-Rojas
PublisherSpringer
Pages49-62
Number of pages14
ISBN (Print)9783030410049
DOIs
StatePublished - 2020
Event6th Latin American High Performance Computing Conference, CARLA 2019 - Turrialba, Costa Rica
Duration: 25 Sep 201927 Sep 2019

Publication series

NameCommunications in Computer and Information Science
Volume1087 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference6th Latin American High Performance Computing Conference, CARLA 2019
Country/TerritoryCosta Rica
CityTurrialba
Period25/09/1927/09/19

Keywords

  • Embedded software
  • High perfomance computing
  • High-speed networks
  • No-SQL

Fingerprint

Dive into the research topics of 'Optimizing Big Data Network Transfers in FPGA SoC Clusters: TECBrain Case Study'. Together they form a unique fingerprint.

Cite this