Journal of Digital Convergence (디지털융복합연구)

The Society of Digital Policy and Management (한국디지털정책학회)
권호 표지
DOI QR코드

DOI QR Code

A SAN Optimization Scheme for High-Performance Storage System

고성능 저장장치를 위한 SAN최적화기법

  • Received : 2013.12.01
  • Accepted : 2014.01.20
  • Published : 2014.01.28
Download PDF
⟨ Previous Next ⟩

Abstract

We noted that substituting hard disk with high-performance storage device on SAN did not immediately result in getting high performance. Investigating the reason behind this leaded us to propose optimization schemes for high-performance storage system. We first got rid of the latency in the I/O process which is unsuitable for the high-performance storage device, added parallelism on the storage server, and applied temporal merge to Superhigh speed network protocol for improving the performance with small random I/O. The proposed scheme was implemented on the SAN with high-performance storage device and we verified that there were about 30% reduction on the I/O delay latency and 200% improvement on the storage bandwidth.

본 논문에서는 SAN(Storage Area Network)에 장착된 하드디스크를 단순히 고성능 저장장치로 교체하면 고성능 저장장치의 뛰어난 성능을 제대로 활용할 수 없음을 확인하고 원인 분석을 하여 고성능 저장장치를 위한 SAN의 성능 최적화기법들을 제안하였다. 먼저 고성능 저장장치에 맞지 않는 기존의 SAN 입출력경로에 존재하던 지연을 없애고, 저장장치 서버에서 입출력 요청들을 병렬 처리할 수 있게 하고, 소형 랜덤 입출력처리의 성능 향상을 위해 SAN에 연결된 초고속 네트워크에 사용되는 기존의 전송 프로토콜에 시간적 병합 기법을 추가하였다. 제안한 기법들의 우수성을 입증하는 방법으로 실제로 고성능 저장장치를 장착한 SAN에 최적화기법들을 구현하였으며, 다양한 입출력 데이터로 실험한 결과 30%이상의 입출력 지연시간 절감과 200%이상의 성능 향상을 확인하였다.

Keywords

References

  1. Ashish Palekar, Design and Implementation of ALINUX SCSI Target for Storage Area Networks, NETWORKS, In ALS' 01
  2. Bob Woodruff, Introduction to the InfiniBand Core Software, Linux Symposium, Vol. 2, pp. 271-282, July 2005.
  3. Burr, G. W., Overview of candidate device technologies for storage-class memory, IBM Journal of Research and Development, Vol. 52, Issue 4.5, pp. 449-464, July 2008. https://doi.org/10.1147/rd.524.0449
  4. Caufield, A. M., Et Al., Moneta: A highperformance storage array architecture for nextgeneration, non-volatile memories, In MICRO'10 (2010), pp. 385-395.
  5. Charawi., S., Using a shared storage class memory device to improve the reliability of RAID arrays, In PDSW 2010, pp.1-5.
  6. Jose B. Cruz, JR, Leader-follower strategies for multilevel systems, IEEE Transactions on Automatic Control, Vol. 23, No. 2, pp. 244 - 255, Apr. 1978. https://doi.org/10.1109/TAC.1978.1101716
  7. DP Bovet, M Cesati, Understanding the Linux Kernel, 3rd Edition, O'REiLLY, 2005.
  8. Freitas, R. F., storage-class memory: The next storage system technology, IBM Journal of Research and Development, Vol. 52, Issue 4.5, pp.439-447, July 2008. https://doi.org/10.1147/rd.524.0439
  9. Fusion-io, ioDrive, http://www.fusionio.com/products/iodrive/
  10. IBM, Introduction to storage Area Networks, http://www.redbooks.ibm.com/redbooks/pdfs/sg245470.pdf
  11. Jens Axboe, Flexible IO Tester, http://git.kernel.dk/?p=fio.git
  12. John Ousterhout, The Case for RAMClouds: Scalable High-Performance storage Entirely in DRAM, ACM SIGOPS Operating Systems Review, Vol. 43, No. 4, pp.92-105, January 2010. https://doi.org/10.1145/1713254.1713276
  13. Mellanox, Building a Scalable Storage with InfiniBand, http://www.mellanox.com/relateddocs/whitepapers/WP_Scalable_Storage_InfiniBand_Final.pdf
  14. Mellanox, InfiniBand storage, http://www.mellanox.com/pdf/whitepapers/I- nfiniBand_storage_WP_050.pdf
  15. Mellanox, Introduction to InfiniBand, http://www.mellanox.com/pdf/whitep-apers/IB_Intro_WP_190.pdf
  16. Mike Ko, Technical Overview of iSCSI Extensions for RDMA (iSER) & Datamover Architecture for iSCSI (DA), RDMA Consortium, 2003.
  17. RAMCloud Project, https://ramcloud.stanford.edu/wiki/display/ramcloud/R-AMCloud
  18. Ru Fang, High Performance Database Logging using Storage Class Memory, In ICDE 2011, pp.1221-1231.
  19. SAP, SAP HANA, http://www.sap.com/solutions/technology/in-memory-computing-platform/index.epx
  20. SCST, Generic SCSI Target Subsystem for Linux, http://scst.sourceforge.net/
  21. TAEJININFOTECH, HHA 3804, http://www.taejin.co.kr
  22. Young Jin Yu, el., Exploiting Peak Device Throughput from Random Access Workload, In Hot storage'12, USENIX.