Efficient Parallel Graph Exploration on Multi-Core CPU and GPU
By Sungpack Hong, Tayo Oguntebi, Kunle Olukotun. International Conference on Parallel Architectures and Compilation Techniques, 2011
294 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/a9d1eda1cafb18b14286815d6c2a60ff6918c837
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
- Krste Asanovic
- David A. Patterson
- Scott Beamer
- Onur Mutlu
- Sungjoo Yoo
- Junwhan Ahn
- Kiyoung Choi
- David M. Brooks
In Computer Vision and Pattern Recognition
- David A. Patterson
- Krste Asanovic
- Scott Beamer
- David M. Brooks
- Onur Mutlu
- Gu-Yeon Wei
- Sungjoo Yoo
- Junwhan Ahn
Citations in the corpus
- GraphPIM: Enabling Instruction-Level PIM Offloading in Graph Computing Frameworks (2017, methodology): “Data Components: Figure 3 illustrates a code snippet of breadth-first search (BFS) using a vertex-frontier based algorithm [18].”
- A scalable processing-in-memory accelerator for parallel graph processing (2015, methodology): “Some works use GPGPUs to accelerate graph processing [15, 18, 19, 40].”
- Scalable GPU graph traversal (2012, background): “While quadratic-work methods might be acceptable in certain very narrow regimes [18, 19], they suffer from high overhead and did not prove effective on even the lowest diameter graphs in our experimental corpus.”
- Ligra: a lightweight graph processing framework for shared memory (2013, background): “Recently, GPU systems have been explored for implementing graph traversal problems [20, 35].”
- PIM-enabled instructions: A low-overhead, locality-aware processing-in-memory architecture (2015, methodology): “PageRank (PR) [18] is a well-known algorithm that calculates the importance of vertices in a graph (see Figure 1).”
- X-Stream: edge-centric graph processing using streaming partitions (2013, background): “Sorting [51] + Random Access [33] |V | |E| B log M B min(|V |, |E| M ) − |V |+ |E|”
- MachSuite: Benchmarks for accelerator design and customized architectures (2014, methodology): “The BFS/BULK code uses a brute-force, data parallel method [26], [25] which is typically used on SIMD and vector architectures.”
- Solving Rubik's Cube (2018, methodology): “[2] Sungpack Hong, Tayo Oguntebi, and Kunle Olukotun.”
- Direction-optimizing Breadth-First Search (2012, methodology): “Our hybrid implementation also provides typical speedups of 2 or greater over prior state-of-the-arts for multicore [1, 10, 15] and GPUs [20] when utilizing the same graphs and the same or similar hardware.”