towards on demand i o forwarding in hpc platforms
play

Towards On-Demand I/O Forwarding in HPC Platforms Jean Luca Bez, - PowerPoint PPT Presentation

Towards On-Demand I/O Forwarding in HPC Platforms Jean Luca Bez, Francieli Zanon Boito, Ramon Nou, Alberto Miranda, Toni Cortes, and Philippe O. A. Navaux jean.bez@inf.ufrgs.br PDSW 2020 International Parallel Data Systems Workshop


  1. Towards On-Demand I/O Forwarding in HPC Platforms Jean Luca Bez, Francieli Zanon Boito, Ramon Nou, Alberto Miranda, Toni Cortes, and Philippe O. A. Navaux jean.bez@inf.ufrgs.br PDSW 2020 — International Parallel Data Systems Workshop

  2. INTRODUCTION Agenda The I/O Forwarding Layer ● ● Motivation FORGE The I/O Forwarding Explorer ● ● Forwarding in MareNostrum 4 Forwarding in SDumont ● ● Conclusion 2

  3. INTRODUCTION The I/O Forwarding Layer Compute Nodes Parallel File System Application A Application B Application C Metadata Servers Client Client Client Meta Server 1 Client Client Client Meta Server 2 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Meta Server M Client Client Client Data Server 1 Client Client Client Data Server 2 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Data Server N Application D Application E Application X Data Servers 3

  4. INTRODUCTION The I/O Forwarding Layer Compute Nodes Parallel File System Application A Application B Application C Metadata Servers Client Client Client Meta Server 1 Client Client Client Meta Server 2 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Meta Server M Client Client Client Data Server 1 Client Client Client Data Server 2 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Data Server N Application D Application E Application X Data Servers 4

  5. INTRODUCTION The I/O Forwarding Layer Compute Nodes Forwarding Layer Parallel File System Application A Application B Application C Metadata Servers Client Client Client Meta Server 1 Client Client Client Meta Server 2 ION 1 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Meta Server M ION 2 ION 3 Client Client Client Data Server 1 ● ● ● ION K Client Client Client Data Server 2 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Data Server N Application D Application E Application X Data Servers 5

  6. INTRODUCTION The I/O Forwarding Layer Compute Nodes Forwarding Layer Parallel File System Application A Application B Application C Metadata Servers Client Client Client Meta Server 1 Client Client Client Meta Server 2 ION 1 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Meta Server M ION 2 ION 3 Client Client Client Data Server 1 ● ● ● ION K Client Client Client Data Server 2 ● ● ● ● ● ● ● ● ● ● ● ● Client Client Client Data Server N Application D Application E Application X Data Servers 6

  7. INTRODUCTION Motivation ● Investigate the impact of I/O forwarding on performance Take into account the application’s access pattern ● ● Most machines cannot be easily reconfigured End-users are not allowed to change this layer ● ● We need a research/exploration alternative! ● When forwarding is the best choice? How many I/O nodes should an application use? ● 7

  8. ARCHITECTURE FORGE: The I/O FORwardinG Explorer Compute Node Compute Node Compute Node Compute Node 2 6 3 7 4 8 5 9 11 15 12 16 10 14 13 16 FORGE CN FORGE CN FORGE CN FORGE CN Compute FORGE FORGE Compute 0 1 Node ION ION Node Parallel File System MPI Rank 8

  9. FORGE EXPERIMENTS MareNostrum 4 (Spain) and Santos Dumont (Brazil) supercomputers ● ● 189 distinct scenarios (access patterns and deployments): Compute nodes: 8, 16, and 32 ○ ○ Client processes per compute node: 12, 24, and 48 (96, 192, 384, 768, and 1536 processes in total) ○ File layout: file-per-process or shared file Spatiality: contiguous or 1D-strided ○ ○ Operation: WRITE Request sizes: 32KB, 128KB, 512KB, 1MB, 4MB, 6MB, and 8MB ○ ○ Stonewall: one second 9

  10. I/O FORWARDING MareNostrum 4 ● Bandwidth at client-side ● 5 repetitions for each ● Different days and periods 10

  11. I/O FORWARDING MareNostrum 4 ● Bandwidth at client-side ● 5 repetitions for each ● Different days and periods 11

  12. I/O FORWARDING MareNostrum 4 ● Bandwidth at client-side ● 5 repetitions for each ● Different days and periods 12

  13. I/O FORWARDING MareNostrum 4 ● Bandwidth at client-side ● 5 repetitions for each ● Different days and periods 13

  14. I/O FORWARDING MareNostrum 4 ● Bandwidth at client-side ● 5 repetitions for each ● Different days and periods 14

  15. I/O FORWARDING MareNostrum 4 ● Bandwidth at client-side ● 5 repetitions for each ● Different days and periods 15

  16. I/O FORWARDING MareNostrum 4 ● How many choices do we have to consider? Dunn’s nonparametric test ● ● 3 choices impact performance 46% patterns (88 out of 189) ● What is the best number of I/O nodes? No simple rule to fit all ● 16

  17. I/O FORWARDING Santos Dumont ● Forwarding impact is different! ● The more I/O nodes, the better ● Not forwarding is an option 17

  18. I/O FORWARDING Santos Dumont ● Forwarding impact is different! ● The more I/O nodes, the better ● Not forwarding is an option 18

  19. I/O FORWARDING Santos Dumont ● Forwarding impact is different! ● The more I/O nodes, the better ● Not forwarding is an option 19

  20. I/O FORWARDING Santos Dumont ● Forwarding impact is different! ● The more I/O nodes, the better ● Not forwarding is an option 20

  21. I/O FORWARDING Santos Dumont ● Forwarding impact is different! ● The more I/O nodes, the better ● Not forwarding is an option 21

  22. I/O FORWARDING Santos Dumont ● Forwarding impact is different! ● The more I/O nodes, the better ● Not forwarding is an option 22

  23. RESULTS Discussion Increasing heterogeneous applications ● ● Shift from must-use to on-demand I/O forwarding layer Transparently reshape the flow of requests ● ● Towards a dynamic allocation of I/O nodes Idle or reserved set of compute nodes could act as I/O nodes ● ● Interference on I/O could not be reduced or eliminated 23

  24. PRESENTATION Conclusion ● I/O forwarding is an established and widely-adopted technique ● Not always possible to explore its advantages under different setups ● Impact or disrupt production systems ● FORGE: a lightweight forwarding layer in user-space ● Understand the impact of forwarding different access patterns ● Evaluation in MareNostrum 4 and Santos Dumont supercomputers ● Shift from must-use to on-demand I/O forwarding layer 24

  25. ACKNOWLEDGMENTS This study was financed by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001. It has also received support from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil; It is also partially supported by the Spanish Ministry of Economy and Competitiveness (MINECO) under grants PID2019-107255GB; and the Generalitat de Catalunya under contract 2014—SGR—1051. The author thankfully acknowledges the computer resources, technical expertise and assistance provided by the Barcelona Supercomputing Center - Centro Nacional de Supercomputación. The authors acknowledge the National Laboratory for Scientific Computing (LNCC/MCTI, Brazil) for providing HPC resources of the SDumont supercomputer, which have contributed to the research results reported within this paper. URL: http://sdumont.lncc.br. 25

  26. Towards On-Demand I/O Forwarding in HPC Platforms Jean Luca Bez, Francieli Zanon Boito, Ramon Nou, Alberto Miranda, Toni Cortes, and Philippe O. A. Navaux jean.bez@inf.ufrgs.br PDSW 2020 — International Parallel Data Systems Workshop

Recommend


More recommend