{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T02:17:08Z","timestamp":1773886628506,"version":"3.50.1"},"reference-count":40,"publisher":"Association for Computing Machinery (ACM)","issue":"14","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2016,10]]},"abstract":"<jats:p>As data sets grow and conventional processor performance scaling slows, data analytics move towards heterogeneous architectures that incorporate hardware accelerators (notably GPUs) to continue scaling performance. However, existing GPU-based databases fail to deal with big data applications efficiently: their execution model suffers from scalability limitations on GPUs whose memory capacity is limited; existing systems fail to consider the discrepancy between fast GPUs and slow storage, which can counteract the benefit of GPU accelerators.<\/jats:p>\n          <jats:p>In this paper, we propose HippogriffDB, an efficient, scalable GPU-accelerated OLAP system. It tackles the bandwidth discrepancy using compression and an optimized data transfer path. HippogriffDB stores tables in a compressed format and uses the GPU for decompression, trading GPU cycles for the improved I\/O bandwidth. To improve the data transfer efficiency, HippogriffDB introduces a peer-to-peer, multi-threaded data transfer mechanism, directly transferring data from the SSD to the GPU. HippogriffDB adopts a query-over-block execution model that provides scalability using a stream-based approach. The model improves kernel efficiency with the operator fusion and double buffering mechanism.<\/jats:p>\n          <jats:p>We have implemented HippogriffDB using an NVMe SSD, which talks directly to a commercial GPU. Results on two popular benchmarks demonstrate its scalability and efficiency. HippogriffDB outperforms existing GPU-based databases (YDB) and in-memory data analytics (MonetDB) by 1-2 orders of magnitude.<\/jats:p>","DOI":"10.14778\/3007328.3007331","type":"journal-article","created":{"date-parts":[[2016,11,1]],"date-time":"2016-11-01T13:47:47Z","timestamp":1478008067000},"page":"1647-1658","source":"Crossref","is-referenced-by-count":67,"title":["HippogriffDB"],"prefix":"10.14778","volume":"9","author":[{"given":"Jing","family":"Li","sequence":"first","affiliation":[{"name":"University of California"}]},{"given":"Hung-Wei","family":"Tseng","sequence":"additional","affiliation":[{"name":"University of California"}]},{"given":"Chunbin","family":"Lin","sequence":"additional","affiliation":[{"name":"University of California"}]},{"given":"Yannis","family":"Papakonstantinou","sequence":"additional","affiliation":[{"name":"University of California"}]},{"given":"Steven","family":"Swanson","sequence":"additional","affiliation":[{"name":"University of California"}]}],"member":"320","published-online":{"date-parts":[[2016,10]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"http:\/\/www.intel.com\/content\/dam\/www\/public\/us\/en\/documents\/product-specifications\/ssd-dc-s3700-spec.pdf.  http:\/\/www.intel.com\/content\/dam\/www\/public\/us\/en\/documents\/product-specifications\/ssd-dc-s3700-spec.pdf."},{"key":"e_1_2_1_2_1","unstructured":"http:\/\/www.nvidia.com\/object\/tesla-servers.html.  http:\/\/www.nvidia.com\/object\/tesla-servers.html."},{"key":"e_1_2_1_3_1","unstructured":"https:\/\/developer.nvidia.com\/gpudirect.  https:\/\/developer.nvidia.com\/gpudirect."},{"key":"e_1_2_1_4_1","unstructured":"http:\/\/blog.pmcs.com\/project-donard-peer-to-peer-communication-with-nvm-express-devices-part-two.  http:\/\/blog.pmcs.com\/project-donard-peer-to-peer-communication-with-nvm-express-devices-part-two."},{"key":"e_1_2_1_5_1","unstructured":"https:\/\/trademarks.justia.com\/865\/43\/nvmedirect-86543720.html.  https:\/\/trademarks.justia.com\/865\/43\/nvmedirect-86543720.html."},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/1376616.1376712"},{"key":"e_1_2_1_8_1","volume-title":"Challenges and opportunities with big data 2011--1","author":"Agrawal D.","year":"2011"},{"key":"e_1_2_1_9_1","first-page":"225","volume-title":"CIDR","volume":"5","author":"Boncz P. A.","year":"2005"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.14778\/2536274.2536325"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2010.36"},{"issue":"5","key":"e_1_2_1_12_1","first-page":"256","article-title":"Design of ion-implanted mosfet's with very small physical dimensions. Solid-State Circuits","volume":"9","author":"Dennard R. H.","year":"1974","journal-title":"IEEE Journal of"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/2000064.2000108"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.14778\/1920841.1920927"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/1142473.1142511"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2011.77"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/1376616.1376670"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.14778\/1952376.1952381"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.14778\/2536206.2536216"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.14778\/2536360.2536370"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/2611567"},{"key":"e_1_2_1_22_1","first-page":"6","volume-title":"OSDI","author":"Kim S.","year":"2014"},{"key":"e_1_2_1_23_1","volume-title":"The data warehouse toolkit: The definitive guide to dimensional modeling","author":"Kimball R.","year":"2013"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.14778\/3007328.3007331"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD.2016.7753307"},{"key":"e_1_2_1_26_1","volume-title":"Knapsack problems: algorithms and computer implementations","author":"Martello S.","year":"1990"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/1964179.1964189"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-10424-4_17"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-44676-1_10"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/InPar.2012.6339599"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/1559845.1559865"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/2370816.2370864"},{"key":"e_1_2_1_33_1","first-page":"67","volume-title":"OSDI","author":"Seshadri S.","year":"2014"},{"key":"e_1_2_1_34_1","volume-title":"A survey of compressed domain processing techniques","author":"Smith B.","year":"1995"},{"key":"e_1_2_1_35_1","volume-title":"A compression method for clustered bit-vectors. Information processing letters, 7(6):308--311","author":"Teuhola J.","year":"1978"},{"key":"e_1_2_1_36_1","unstructured":"H.-W. Tseng Y. Liu M. Gahagan J. Li Y. Jin and S. Swanson. Gullfoss: Accelerating and simplifying data movement among heterogeneous computing and storage resources. Technical report.  H.-W. Tseng Y. Liu M. Gahagan J. Li Y. Jin and S. Swanson. Gullfoss: Accelerating and simplifying data movement among heterogeneous computing and storage resources. Technical report."},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/2749469.2750399"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.14778\/2350229.2350268"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.14778\/2732967.2732976"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2012.19"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.14778\/2536206.2536210"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3007328.3007331","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,28]],"date-time":"2022-12-28T11:08:33Z","timestamp":1672225713000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3007328.3007331"}},"subtitle":["balancing I\/O and GPU bandwidth in big data analytics"],"short-title":[],"issued":{"date-parts":[[2016,10]]},"references-count":40,"journal-issue":{"issue":"14","published-print":{"date-parts":[[2016,10]]}},"alternative-id":["10.14778\/3007328.3007331"],"URL":"https:\/\/doi.org\/10.14778\/3007328.3007331","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2016,10]]}}}