Merge: a programming model for heterogeneous multi-core systems

By Michael D. Linderman, Jamison D. Collins, Hong Wang, Teresa Huai-Ying Meng. International Conference on Architectural Support for Programming Languages and Operating Systems, 2008

330 citations recorded by the source corpus.

How researchers see this paper

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

Qilin: Exploiting parallelism on heterogeneous multiprocessors with adaptive mapping (2009, methodology): “RapidMind [29], Cell [26, 5, 22], Merge [16], and Qilin target both CPU and GPU.”
Mars: A MapReduce Framework on graphics processors (2008, methodology): “[20] proposed the Merge framework for dynamically scheduling MapReduce tasks among heterogenous processors.”
An analytical model for a GPU architecture with memory-level and thread-level parallelism awareness (2009, methodology): “MWP represents the maximum number of warps per SM that can access the memory simultaneously during the time period from right after the SM processor executes a memory instruction from one warp (therefore, memory requests are just 200 0 4 9 16 25 36 49 64 81 100 121 144 169 196 225 256 289 324 361”
Dandelion: a compiler and runtime for heterogeneous systems (2013, background): “Merge [66] provides a high-level language based on MapReduce that transparently chooses amongst implementations dynamically, it is targeted at a single machine unlike Dandelion.”
An integrated GPU power and performance model (2010, methodology): “306 Sepia [15] Filter for artificially aging images 12.”