A spatial index for hybrid storage

The introduction of flash SSDs has accelerated the performance of DBMSes. However, the intrinsic characteristics of flash motivated many researchers to investigate new efficient data structures. The emergence of 3DXPoint, a new non-volatile memory, sets new challenges: 3DXPoint features low latency and high IOPS even at small queue depths. However, the cost of 3DXPoint is 4 times higher than that of a flash-based device, rendering hybrid storage systems a good alternative. In this paper we pursue exploiting the efficiency of both 3DXPoint and flash-based devices introducing H-Grid, a variant of Grid-File for hybrid storage. H-Grid uses a flash SSD as main store and a small 3DXPoint device to persist the hottest data. The performance of the proposed index is experimentally evaluated, comparing it against GFFM, a flash efficient implementation of Grid File. The results show that H-Grid is faster than GFFM execution on a flash SSD, reducing the single point search time from 35% up to 43%.

[1]  Yi Liu,et al.  MOLAR: A cost-efficient, high-performance hybrid storage cache , 2013, 2013 IEEE International Conference on Cluster Computing (CLUSTER).

[2]  Bingsheng He,et al.  Tree Indexing on Flash Disks , 2009, 2009 IEEE 25th International Conference on Data Engineering.

[3]  Athanasios Fevgas,et al.  Grid-File: Towards to a Flash Efficient Multi-dimensional Index , 2015, DEXA.

[4]  John L. Gustafson,et al.  Little's Law , 2011, Encyclopedia of Parallel Computing.

[5]  Chong Leong Gan,et al.  3D Flash Memories , 2016, Microelectron. Reliab..

[6]  Peiquan Jin,et al.  XB+-Tree: A Novel Index for PCM/DRAM-Based Hybrid Memory , 2016, ADC.

[7]  YueLihua,et al.  Optimizing B+-tree for hybrid storage systems , 2015 .

[8]  Sang-Won Lee,et al.  B+-tree Index Optimization by Exploiting Internal Parallelism of Flash-based Solid State Drives , 2011, Proc. VLDB Endow..

[9]  Peiquan Jin,et al.  Optimizing R-tree for flash memory , 2015, Expert Syst. Appl..

[10]  Athanasios Fevgas,et al.  LB-Grid: An SSD efficient Grid File , 2019, Data Knowl. Eng..

[11]  Jürg Nievergelt,et al.  The Grid File: An Adaptable, Symmetric Multikey File Structure , 1984, TODS.

[12]  Athanasios Fevgas,et al.  Bulk-loading and bulk-insertion algorithms for $$\hbox {xBR}^{+}\hbox {-trees}$$xBR+-trees in Solid State Drives , 2019, Computing.

[13]  Jun Xu,et al.  Hybrid Storage Systems: A Survey of Architectures and Algorithms , 2018, IEEE Access.

[14]  Kenneth Salem,et al.  Hybrid Storage Management for Database Systems , 2013, Proc. VLDB Endow..

[15]  Kenneth A. Ross,et al.  SSD bufferpool extensions for database systems , 2010, Proc. VLDB Endow..

[16]  Dimitrios Gunopulos,et al.  Efficient indexing data structures for flash-based sensor devices , 2006, TOS.

[17]  Dan Williams,et al.  Platform Storage Performance With 3D XPoint Technology , 2017, Proceedings of the IEEE.

[18]  Ramesh K. Sitaraman,et al.  Lazy-Adaptive Tree: An Optimized Index Structure for Flash Devices , 2009, Proc. VLDB Endow..

[19]  Christian S. Jensen,et al.  Read/write-optimized tree indexing for solid-state drives , 2015, The VLDB Journal.

[20]  Guangyu Sun,et al.  A Probabilistic Data Replacement Strategy for Flash-Based Hybrid Storage System , 2013, APWeb.

[21]  Ling Yuan,et al.  Efficient implementation of a multi-dimensional index structure over flash memory storage systems , 2011, The Journal of Supercomputing.

[22]  David J. Lilja,et al.  Reducing Relational Database Performance Bottlenecks Using 3D XPoint Storage Technology , 2018, 2018 17th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/ 12th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE).

[23]  Y. Manolopoulos,et al.  Bulk-loading and bulk-insertion algorithms for xBR +-trees in Solid State Drives , 2019 .

[24]  Klaus H. Hinrichs,et al.  Implementation of the grid file: Design concepts and experience , 1985, BIT.

[25]  Ioannis Koltsidas.  IBM Storage and the NVM Express Revolution , 2017 .

[26]  Hanan Samet,et al.  Hierarchical Spatial Data Structures , 1989, SSD.

[27]  Peiquan Jin,et al.  Optimizing B+-tree for hybrid storage systems , 2014, Distributed and Parallel Databases.

[28]  Sang-Won Lee,et al.  MPSearch: Multi-Path Search for Tree-based Indexes to Exploit Internal Parallelism of Flash SSDs , 2014, IEEE Data Eng. Bull..

[29]  Mahmut T. Kandemir,et al.  FlashShare: Punching Through Server Storage Stack from Kernel to Firmware for Ultra-Low Latency SSDs , 2018, OSDI.

[30]  Suman Nath,et al.  FAST: A Generic Framework for Flash-Aware Spatial Trees , 2011, SSTD.

[31]  Hanan Samet,et al.  Applications of spatial data structures , 1989 .

[32]  Hanan Samet,et al.  Applications of spatial data structures - computer graphics, image processing, and GIS , 1990 .

[33]  Sungho Kim,et al.  AS B-tree: A Study of an Efficient B+-tree for SSDs , 2014, J. Inf. Sci. Eng..

[34]  Feng Chen,et al.  Hystor: making the best use of solid state drives in high performance storage systems , 2011, ICS '11.

[35]  Cristina Dutra de Aguiar Ciferri,et al.  A Generic and Efficient Framework for Spatial Indexing on Flash-Based Solid State Drives , 2017, ADBIS.