can microservices drive a renaissance in workload aware
play

Can Microservices Drive a Renaissance in Workload-Aware Storage - PowerPoint PPT Presentation

Can Microservices Drive a Renaissance in Workload-Aware Storage Management? 12th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 20) Pranav Bhandari 1 , Lukas Rupprecht 2 , Dimitrios Skourtis 2 , Ali Anwar 2 , Deepavali


  1. Can Microservices Drive a Renaissance in Workload-Aware Storage Management? 12th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage ’20) Pranav Bhandari 1 , Lukas Rupprecht 2 , Dimitrios Skourtis 2 , Ali Anwar 2 , Deepavali Bhagwat 2 , Vasily Tarasov 2 , Avani Wildani 1 1 Emory University, 2 IBM 1

  2. Microservices ➔ Isolation ➔ Flexibility ➔ Productivity ➔ Scalability ➔ Storage? Gan, Yu, et al. "An open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems." Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems 2 (ASPLOS) . 2019.

  3. Storage APP A B C STORAGE Distributed Storage (OpenEBS, Ceph, GPFS, GlusterFS) 3

  4. Research Monolithic Workload Characterization Microservices A B C Distributed Storage Auto-tuning 4

  5. Motivation Monolithic Microservices A heterogeneous A set of workloads mix of workloads from application Microservices Monolithic from different components application (services) that each A B C components perform a simple task DFS Shared storage Each microservice can have its own volume which can be provisioned dynamically 5

  6. Workload Stability ➔ Access pattern based workload metrics ◆ read/write ratio ◆ locality ◆ I/O size distribution ➔ Are these metrics more stable in the workloads of microservices compared to the monolithic workload of functionally similar application? 6

  7. Storage Auto-tuning ➔ Storage parameters ◆ cache (size, write policy, replacement policy, prefetching) ◆ replication ◆ block size ➔ Case study: Cache Size 7

  8. Setting ➔ Isolated I/O cache in host memory ◆ Local in-memory data access ◆ No network request to the storage service ➔ Cache allocated per persistent volume mounted on the host ◆ Size cache based on the workload of the persistent volume 8

  9. Shared vs Isolated Cache Isolated cache performs better than shared cache! 9

  10. Cache Size Allocation Workload analysis is needed for cache allocation! 10

  11. Initial Design 11

  12. Thank You! Can Microservices Drive a Renaissance in Workload-Aware Storage Management? Pranav Bhandari 1 , Lukas Rupprecht 2 , Dimitrios Skourtis 2 , Ali Anwar 2 , Deepavali Bhagwat 2 , Vasily Tarasov 2 , Avani Wildani 1 1 Emory University, 2 IBM pranav.bhandari@emory.edu 12

Recommend


More recommend