ABSTRACT
This paper presents a heterogeneous database hardware accelerator MPSoC manufactured in 28 nm SLP CMOS. The 18 mm2 chip integrates a runtime task scheduling unit for energy-efficient query processing and hierarchical power management supported by an ultra-fast dynamic voltage and frequency scaling. Four processing elements, connected by a star-mesh network-on-chip, are accelerated by an instruction set extension tailored to fundamental data-intensive applications. We evaluate the MPSoC with typical database benchmarks focusing on scans and bitmap operations. When the processing elements operate on data stored in local memories, the chip consumes 250 mW and shows a 96x energy efficiency improvement compared to state-of-the-art platforms.
- I. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci. A survey on sensor networks. IEEE Communications Magazine, 40(8):102--114, Aug 2002. Google ScholarDigital Library
- O. Arnold, S. Haas, G. Fettweis, B. Schlegel, T. Kissinger, T. Karnagel, and W. Lehner. Hashi: An application-specific instruction set extension for hashing. In ADMS, pages 25--33, 2014.Google Scholar
- O. Arnold, S. Haas, G. Fettweis, B. Schlegel, T. Kissinger, and W. Lehner. An application-specific instruction set for accelerating set-oriented database primitives. In ACM SIGMOD, pages 767--778, June 2014. Google ScholarDigital Library
- O. Arnold, E. Matus, B. Noethen, M. Winter, T. Limberg, and G. Fettweis. Tomahawk: Parallelism and heterogeneity in communications signal processing mpsocs. ACM Trans. Embed. Comput. Syst., 13(3s):107:1--107:24, 2014. Google ScholarDigital Library
- J. Chhugani, A. D. Nguyen, V. W. Lee, W. Macy, M. Hagog, Y.-K. Chen, A. Baransi, S. Kumar, and P. Dubey. efficient implementation of sorting on multi-core simd cpu architecture. Proc. VLDB Endow., 1(2):1313--1324, 2008. Google ScholarDigital Library
- F. Fusco, M. Vlachos, X. Dimitropoulos, and L. Deri. Indexing million of packets per second using gpus. IMC'13, pages 327--332, October 2013. Google ScholarDigital Library
- B. He, M. Lu, K. Yang, R. Fang, N. K. Govindaraju, Q. Luo, and P. V. Sander. Relational query coprocessing on graphics processors. ACM Trans. Database Syst., 34(4):21:1--21:39, 2009. Google ScholarDigital Library
- S. Höppner, S. Hänzsche, G. Ellguth, D. Walter, H. Eisenreich, and R. Schüffny. A fast-locking adpll with instantaneous restart capability in 28-nm cmos technology. IEEE Transactions on Circuits and Systems II: Express Briefs, 60(11):741--745, Nov 2013.Google ScholarCross Ref
- S. Höppner, C. Shao, H. Eisenreich, G. Ellguth, M. Ander, and R. Schüffny. A power management architecture for fast per-core dvfs in heterogeneous mpsocs. In IEEE International Symposium on Circuits and Systems (ISCAS), pages 261--264, May 2012.Google ScholarCross Ref
- S. Höppner, D. Walter, T. Hocker, S. Henker, S. Hänzsche, D. Sausner, G. Ellguth, J.-U. Schlüßler, et al. An energy efficient multi-gbit/s noc transceiver architecture with combined ac/dc drivers and stoppable clocking in 65 nm and 28 nm cmos. IEEE Journal of Solid-State Circuits, 50(3):749--762, March 2015.Google ScholarCross Ref
- S. Kanev, J. P. Darago, and K. Hazelwood. Profiling a warehouse-scale computer. In Proceedings of the 42nd International Symposium on Computer Architecture, ISCA'15, June 2015. Google ScholarDigital Library
- P. Li, J. L. Shin, G. Konstadindis, F. Schumacher, V. Krishnaswamy, H. Cho, S. Dash, R. Masleid, et al. A 20nm 32-core 64mb l3 cache sparc m7 processor. In ISSCC Dig. Tech. Papers, ISSCC'15, pages 72--73, Feb 2015.Google Scholar
- M. Mazzola, G. Schaaf, F. Niewels, and T. Kurner. Exploration of centralized car2x-systems over lte. In 2015 IEEE 81st Vehicular Technology Conference (VTC Spring), pages 1--5, May 2015.Google ScholarCross Ref
- B. Nöthen, O. Arnold, E. P. Adeva, T. Seifert, E. Fischer, S. Kunze, E. Matus, and G. F. others. A 105gops 36mm2 heterogeneous sdr mpsoc with energy-aware dynamic scheduling and iterative detection-decoding for 4g in 65nm cmos. In ISSCC Dig. Tech. Papers, ISSCC'14, pages 188--189, Feb 2014.Google Scholar
- I. Psaroudakis, T. Kissinger, D. Porobic, T. Ilsche, E. Liarou, P. Tözün, A. Ailamaki, and W. Lehner. Dynamic fine-grained scheduling for energy-efficient main-memory queries. DaMoN'14, pages 1:1--1:7. ACM, 2014. Google ScholarDigital Library
- M. Satyanarayanan, R. Schuster, M. Ebling, G. Fettweis, H. Flinck, K. Joshi, and K. Sabnani. An open ecosystem for mobile-cloud convergence. In IEEE Communications Magazine, vol. 53, no. 3, pages 63--70, March 2015.Google Scholar
- B. Schlegel, T. Willhalm, and W. Lehner. Fast sorted-set intersection using simd instructions. ADMS'11, 2011.Google Scholar
- B. Sukhwani, H. Min, M. Thoennes, P. Dube, B. Iyer, B. Brezzo, D. Dillenberger, and S. Asaad. Database analytics acceleration using fpgas. PACT'12, pages 411--420, 2012. Google ScholarDigital Library
- D. Tsirogiannis, S. Harizopoulos, and M. A. Shah. Analyzing the energy efficiency of a database server. SIGMOD'10, pages 231--242. ACM, 2010. Google ScholarDigital Library
- A. Ungethüm, D. Habich, T. Karnagel, W. Lehner, N. Asmussen, M. Völp, B. Nöthen, and G. Fettweis. Query processing on low-energy many-core processors. HardBD'15, 2015.Google ScholarCross Ref
- K. Wu, E. J. Otoo, and A. Shoshani. An efficient compression scheme for bitmap indices. Technical report, ACM Transactions on Database Systems, 2004.Google Scholar
- L. Wu, A. Lottarini, T. K. Paine, M. A. Kim, and K. A. Ross. Q100: The architecture and design of a database processing unit. ASPLOS'14, pages 255--268. ACM, 2014. Google ScholarDigital Library
Recommendations
Efficient Query Processing on Many-core Architectures: A Case Study with Intel Xeon Phi Processor
SIGMOD '16: Proceedings of the 2016 International Conference on Management of DataRecently, Intel Xeon Phi is emerging as a many-core processor with up to 61 x86 cores. In this demonstration, we present PhiDB, an OLAP query processor with simultaneous multi-threading (SMT) capabilities on Xeon Phi as a case study for parallel ...
Energy-Efficient Query Processing on Embedded CPU-GPU Architectures
DaMoN'15: Proceedings of the 11th International Workshop on Data Management on New HardwareEnergy efficiency is a major design and optimization factor for query co-processing of databases in embedded devices. Recently, GPUs of new-generation embedded devices have evolved with the programmability and computational capability for general-...
Comments