First Investigations of the Vector Supercomputer SX-Aurora TSUBASA as a Co-Processor for Database Systems

The hardware landscape is currently changing from homogeneous multi-core systems towards heterogeneous systems with many different computing units, each with their own characteristics. This trend is a great opportunity for database systems to increase the overall performance if the heterogeneous resources can be utilized efficiently. Following that trend, NEC cooperation has recently introduced a novel heterogeneous hardware system called SX-Aurora TSUBASA. This novel heterogeneous system features a strong vector engine as a (co-)processor providing world’s highest memory bandwidth of 1.2TB/s per vector processor. From a database system perspective, where many operations are memory bound, this bandwidth is very interesting. Thus, we describe the unique architecture and properties of this novel heterogeneous system in this paper. Moreover, we present first database-specific evaluation results to show the benefit of this system to increase the query performance. We conclude the paper with an outlook on our ongoing research activities in this direction.

[1]  Jignesh M. Patel,et al.  BitWeaving: fast scans for main memory data processing , 2013, SIGMOD '13.

[2]  Hiroaki Kobayashi,et al.  Performance Evaluation of a Vector Supercomputer SX-Aurora TSUBASA , 2018, SC18: International Conference for High Performance Computing, Networking, Storage and Analysis.

[3]  Wolfgang Lehner,et al.  Column Scan Acceleration in Hybrid CPU-FPGA Systems , 2018, ADMS@VLDB.

[4]  Wolfgang Lehner,et al.  Adaptive Work Placement for Query Processing on Heterogeneous Computing Resources , 2017, Proc. VLDB Endow..

[5]  Ismail Oukid,et al.  Memory Management Techniques for Large-Scale Persistent-Main-Memory Systems , 2017, Proc. VLDB Endow..

[6]  Marcin Zukowski,et al.  Vectorwise: A Vectorized Analytical DBMS , 2012, 2012 IEEE 28th International Conference on Data Engineering.

[7]  Wolfgang Lehner,et al.  ERIS: A NUMA-Aware In-Memory Storage Engine for Analytical Workload , 2014, ADMS@VLDB.

[8]  Bingsheng He,et al.  In-Cache Query Co-Processing on Coupled CPU-GPU Architectures , 2014, Proc. VLDB Endow..

[9]  Guy M. Lohman,et al.  Optimizing GPU-accelerated Group-By and Aggregation , 2015, ADMS@VLDB.

[10]  Wolfgang Lehner,et al.  Lightweight Data Compression Algorithms: An Experimental Survey (Experiments and Analyses) , 2017, EDBT.

[11]  Wolfgang Lehner,et al.  Adaptive Energy-Control for In-Memory Database Systems , 2018, SIGMOD Conference.

[12]  Kenneth A. Ross,et al.  Rethinking SIMD Vectorization for In-Memory Databases , 2015, SIGMOD Conference.

[13]  Feng Li,et al.  Accelerating Relational Databases by Leveraging Remote Memory and RDMA , 2016, SIGMOD Conference.

[14]  Martin L. Kersten,et al.  MonetDB: Two Decades of Research in Column-oriented Database Architectures , 2012, IEEE Data Eng. Bull..

[15]  Martin L. Kersten,et al.  Breaking the memory wall in MonetDB , 2008, CACM.

[16]  Michael Stonebraker,et al.  C-Store: A Column-oriented DBMS , 2005, VLDB.

[17]  Wolfgang Lehner,et al.  Compression-Aware In-Memory Query Processing: Vision, System Design and Beyond , 2016, ADMS/IMDM@VLDB.

[18]  Daniel J. Abadi,et al.  Integrating compression and execution in column-oriented database systems , 2006, SIGMOD Conference.

[19]  Marcin Zukowski,et al.  Super-Scalar RAM-CPU Cache Compression , 2006, 22nd International Conference on Data Engineering (ICDE'06).

[20]  Karthikeyan Sankaralingam,et al.  Dark Silicon and the End of Multicore Scaling , 2012, IEEE Micro.

[21]  Carsten Binnig,et al.  Dictionary-based order-preserving string compression for main memory column stores , 2009, SIGMOD Conference.

[22]  Wolfgang Lehner,et al.  Diversity of Processing Units , 2018, Datenbank-Spektrum.