RISP: A Reconfigurable In-Storage Processing Framework with Energy-Awareness

@article{Song2018RISPAR,
  title={RISP: A Reconfigurable In-Storage Processing Framework with Energy-Awareness},
  author={Xiaojia Song and Tao Xie and Wen Pan},
  journal={2018 18th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)},
  year={2018},
  pages={193-202}
}
Existing in-storage processing (ISP) techniques mainly focus on maximizing data processing rate by always utilizing total storage data processing resources for all applications. We find that this "always running in full gear" strategy wastes energy for some applications with a low data processing complexity. In this paper we propose RISP (Reconfigurable ISP), an energy-aware reconfigurable ISP framework that employs FPGA as data processing cells and NVM controllers. It can reconfigure storage… CONTINUE READING

Figures, Tables, Results, and Topics from this paper.

Key Quantitative Results

  • Further, its reconfigurability can provide up to 77.2% additional energy saving by judiciously enabling data processing resources that are sufficient for an application.
  • Experimental results show that RISP improves performance by 1.6-25.4× while reducing energy consumption by 2.2-161×. Its reconfigurability offers up to 77.2% additional energy-saving.

References

Publications referenced by this paper.
SHOWING 1-10 OF 20 REFERENCES

MineBench: A Benchmark Suite for Data Mining Workloads

  • 2006 IEEE International Symposium on Workload Characterization
  • 2006
VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL

Summarizer: Trading Communication with Computing Near Storage

  • 2017 50th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO)
  • 2017
VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

BlueDBM: An appliance for Big Data analytics

  • 2015 ACM/IEEE 42nd Annual International Symposium on Computer Architecture (ISCA)
  • 2015
VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

Minerva: Accelerating Data Analysis in Next-Generation SSDs

  • 2013 IEEE 21st Annual International Symposium on Field-Programmable Custom Computing Machines
  • 2013
VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL