GPUTeraSort: high performance graphics co-processor sorting for large database management

By Naga K. Govindaraju, Jim Gray, Ritesh Kumar, Dinesh Manocha. SIGMOD Conference, 2006

496 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/580835ee30534043f53b1cb03fe1f27ce85bcdcf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

Designing efficient sorting algorithms for manycore GPUs (2009, background): “[8] is one of the best performing graphics-based sorts, although it suffers from the O(n log2 n) work complexity typical of bitonic methods.”
Debunking the 100X GPU vs. CPU myth: an evaluation of throughput computing on CPU and GPU (2010, background): “A number of studies [8, 9, 19, 20, 21, 25, 27, 34, 40, 43, 53] discuss similar throughput computing kernels as in this paper.”
Optimizing Data Warehousing Applications for GPUs Using Kernel Fusion/Fission (2012, background): “Further, other groups focus on designing algorithms to accelerate individual operators [1], [20], [21], [22], [23], [24].”
State-of-the-art in heterogeneous computing (2010, methodology): “Sorting Bitonic sort (GPU) [147]; Hybrid radix-bitonic, out-of-core (GPU) [65]; Radix (GPU) [154]; AA-sort (CBEA) [93]”
Where is the data? Why you cannot debate CPU vs. GPU performance without the answer (2011, background): “Additionally, they recognize that in order to benefit from a GPU solution, they must amortize the cost of moving the database to the device by performing multiple database operations.”
JouleSort: a balanced energy-efficiency benchmark (2007, background): “The GPUTeraSort has shown that GPUs can provide much better in-memory sorting bandwidth than CPUs [15].”
Sort vs. Hash Revisited: Fast Join Implementation on Modern Multi-Core CPUs (2009, methodology): “Bitonic sort has also been implemented on GPUs [16, 13, 32] since it mapped well to the high compute density and bandwidth of GPUs.”
Mars: A MapReduce Framework on graphics processors (2008, methodology): “[12] presented novel GPU-based algorithms for the bitonic sort.”
Relational joins on graphics processors (2008, methodology): “presented novel GPUbased algorithms for relational operators including selections, aggregations [15] as well as sorting [14], and for data mining operations such as computing frequencies and quantiles for data”