The CMU 15-418/618 Parallelism Competition gives students the opportunity to win fame, glory, and excellent prizes! It also provides students the chance to perform an extended investigation of a parallel (or high-efficiency) programming task or a parallel system design problem of their choosing.
Competition Winners
- 1st prize: Mosaix - An iOS Photo Mosaic App by Nathan Eliason and Shelly Bensal
- 2nd prize: Accel.js by Leiyu Zhao and Hailiang Xu
- 3rd prize: Multiverse Reinforcement Learning by Bo Chen and Bowen Deng
- 4th prize: CuLSTM: A Python-based DSL for LSTMs on GPU by Yuang Liu and Shuyao Bi
- 5th prize: ParaHMM: Parallel Hidden Markov Model by Yuchen Huo and Danhao Guo
- 6th prize: Particle-based SLAM for Autonomous Vehicles by Nishad Gothoskar and Cyrus Tabrizi
- 7th prize: Flow on the Go by Ashwin Sekar and Richard Zhao
- 8th prize: StandardML-GPU by Rohan Yadav and Brandon Houghton
Full Project List
Systems and Frameworks
- StandardML-GPU by Rohan Yadav and Brandon Houghton
- MiniSpark by Zebing Lin
- FaaSt Py: Distributed Computing Using AWS Lambda by Mrigesh Kalvani and Sandeep Agarwalla
- Serverless Map + Reduce with AWS Lambda by Audasia Ho and Lillian Choung
- Accel.js by Leiyu Zhao and Hailiang Xu
- tScanner: Distributed Video Processing in TensorFlow by Priyanka Kukreja and Matthieu Felix
- paraFS: CloudFS with Parallel Dedup by Pavan Rao and Sanhita Dhavale
- Analyzing Parallel Memory Allocators by Aditya Mathur and Nupur Kulkarni
- Elixir: A Scheduling Framework for Video Processing Pipelines by Tian Xin and Mengjin Yan
- ISPC++ by Aaron Gutierrez
- Python Automatic Loop Parallelization by Joey Fernau
- Flo: A Simple C++ Library for Parallel Max Flow Computations by Joshua Korn
- Workload-Aware Power Management in Jetson Tegra K1 by Chandirasekarendiran Anandan and Gokul Krishnaa Devaraju
- Helium: Neon Math Library by Steven Fu and Christopher Hung
- Optimizing Scanner Performance on PHI clusters by Feng Xiong and Yang Yu
Cache Simulation
- Directory-Based Cache Coherence Protocol by Ridhi Surana and Yinyi Chen
- Cache Coherence Protocols Analyzer by Kshitiz Dange and Yash Tibrewal
Parallel Data Structures
- CStore: A Lock-free SIMD Skip List by Ziyi Liu and Quan Quan
- ParaHash by Brenda Thayillam and Emily Tsui
- ScaleCuckoo by Priya Ranjan Jha and Deep Shah
- Lock-free Cuckoo Hash by Calvin Giroud and Ross Coker
- A Faster Parallel Implementation of a Lock-Free Hash Set by Danny Balter
- Lock Free Concurrent Hashmap by Yangqingwei Shi and Yiqun Ouyang
- Lock Free Binary Search Trees by Vasu Agrawal and Lukas Peraza
- LockTransform by Aishwarya Prem Renu
- PQ Unlocked: Lock-free Priority Queue by Tara Stentz and Omkar Kelkar
- Lock Free Binary Decision Diagram by Feng Xiong
- Creating a Lock-free Linked List Data Structure by Matt Thompson
Graph Algorithms
- ParaGraphL: WebGL-powered Parallel Large-scale Graph Layout Visualization Framework by Tao Lin and Bowei Chen
- Parallel Closeness Centrality with Application to Facebook Graphs by Anton Kuznetsov and Aditya Bhushan
- Parallel Low Diameter Decompositions in Graphs by Jacob Imola and Sidhanth Mohanty
- WMATA Incident Mitigation Strategies by Sunny Zheng
- Parallel Triangle Counting using CUDA by Vashishtha Adtani and Manish Jain
- Parallelizing Boruvka's Algorithm by Jordan Widjaja and Benjamin Chu
Other Algorithms
- Fast Parallel Sort by Kevin Zhang and Matthew Lee
- QoSort: A parallel, Generic Sorting Library in Go by Shaojie Bai and Yutong Chen
- Parallelized Sequential Pattern Mining by Anna Tan
- QuickMatch: Evaluation of Parallel Regular Expression Matching on the GPU by Madhumitha Sridhara and Bharath Kumar M J
- Parallel Lossless Data Compression Algorithms by Zhanxiang Huang and Chen Luo
Computer Vision
- GPU Tracking by Denis Merigoux and Ilai Deutel
- GPU Facial Detection by Brian Jay
- Particle-based SLAM for Autonomous Vehicles by Nishad Gothoskar and Cyrus Tabrizi
- Parallel GPU Image Segmentation by Sandy Jiang and Priscilla Tai
- CU-BRIEF by Abhinit Modi and Luis Fernando Fraga Gonzalez
- Flow on the Go by Ashwin Sekar and Richard Zhao
- Parallel Nearest Neighbor Search for Velodyne LiDAR Point Clouds by Adriel Luo and Amanda Eng
- Real Time Motion Tracking on Raspberry Pi by Tina Luo and Robert Rudolph
- Monocular Visual Odometry on a Jetson Tegra X1 by Sean Buckley and David Zhu
- Optimizing BoofCV Object Tracker on a Smartphone by Shreedutt Hegde and Ashutosh Tadkase
- Fast Parallel Object Tracking by Timothy Li
DNNs on FPGAs
- Convolutional Nets on FPGA by Ricson Cheng and Sree Vishant Prabhakaran
- Accelerating SqueezeNet on FPGA by Samyukta Lanka and Megha Arora
Deep Nets and Machine Learning
- CuLSTM: A Python-based DSL for LSTMs on GPU by Yuang Liu and Shuyao Bi
- Parallel Random Forest by Kirn Hans and Sally McNichols
- DistNet: Distributed Training for DNN by Rucha Sanjay Vaidya and Nasrin Jaleel
- GPU Parallel Best Subset Regression by Benjamin Sauk
- Parallelizing a Deep Neural Network by Gaurav Lahiry and Abhy Vytheeswaran
- LSTM-DSL by Tejus S and Parth Chadha
- Multiverse Reinforcement Learning by Bo Chen and Bowen Deng
- Parallelizing Pre-Training of Deep Neural Networks by Arun Sai Krishnan and Prakruthi Prabhakar
- WisdomLDA by Ye Qi and Yuhan Mao
- Parallel Classification Decision Trees by Abhinand Sukumar and Fadhil Abubaker
- Finding the Optimal Number of Hidden Nodes in a Neural Network by Priscilla Jenq
- CuGB: Parallelizing Gradient Boosting on GPU by Enzhe Lu and Jack Dong
- LSTM Scheduler by Yifan Jiang and Xiangguang Zheng
- Distributed Decision Trees with Heterogeneous Parallelism by Rui Peng and Alex Xiao
- CUDA AdaBoost with Multiple Learners by Rohan Jadvani
- Ask Me Anything: Distributed Question Answering System with Freebase by Ziyuan Gong and Hongyu Li
- ParaHMM: Parallel Hidden Markov Model by Yuchen Huo and Danhao Guo
- DL-Half: A Mini Deep Learning Framework in Halide by Qiaoyu Deng and Changkai Zhou
- Parallel LDA by Chen Hu and Jingcheng Yu
Fast Linear Algebra
- ParGMRES: A Parallel Linear Solver by Yu-Lun Tsai and Chih-Wei Chang
- Sparse Matrix Multiplication by Toshitha Jagadeesh and Anisha Padwekar
- Hyperdrive: A Fast, Parallel Linear Solver by Prerit Rodney and Avisha Dhisle
Graphics and Simulation
- Image Approximation Using a Parallel Genetic Algorithm by Sriharini Pingali and Everi Osofsky
- StitchIt by Sai Harshini Nimmala and Akanksha Periwal
- Mosaix - An iOS Photo Mosaic App by Nathan Eliason and Shelly Bensal
- Parallel Image Similarity Comparison by Jianing Qian and Bryan Lee
- Parallel Point Cloud Processing and Segmentation by Rohan Varma and Ardra Singh
- Parallel GPU Fluid Simulator by Anubhav Jaiswal and Rohit Ramesh Pillai
- CUDA Accelerated Position Based Fluid Simulation by Eric Fang and Connor Maines
- Parallel High-Resolution Fluid Simulation by Jennifer Chou and Sienna Stritter
- Comparing the Performance for Cloth Simulation by Karthic Palaniappan and Isaac Manjarres
- Parallel Galaxy Evolution Simulator by Yunpeng Xu and Zhikun Lao
- Raytracing Sound in 3D Space Using the Microsoft HoloLens by Sara Adkins and Jonathan Merrin
- GPU-Accelerated Source Engine Radiosity Compilation by Ryan Lam and Lucy Tan
- Efficient Ray Tracing on a GPU by Robert Li and Rachel Gu
- GPU Fast Ray Tracing by Cole Heathershaw and Sohail Sidique
- Parallelized Real-Time CPU Raytracing by Jocelyn Huang
- Efficient Ray Tracing by Vivian Wang and Judy Mai
- WorldVu: A 360 Degree Frame Processing Accelerator by Harish Dattatraya Dixit and Sampath Chanda
Games and Planning
- Connect 418 by Elizabeth Ji and Howard Chen
- Chess AI by Sung Jin kim
- Nonogram Solver by Sean Xiao and Nicholas Mu
- ParaSudoku by Maitreyee Joshi and Tingyu Bi
- Parallel Risk AI by Felipe Archondo and Harry Cannon
- Parallel Game Tree Evaluation by Nancy Xiao and Jemmin Chang
- Solving the 15-Puzzle with Parallel A* by Anna Gupta and William Xiao
- PQ Unlocked: Lock-Free Priority Queue for Use in A* by Tara Stentz and Omkar Kelkar
- A Parallelized Algorithm for A* Path Planning by Vivek Sridhar and Colin Wee
- GTO Heads Up No Limit Texas Hold Em Solver by David Steere Connor Watts
Support and Contributions
The 2017 competition was made more fun with prizes donated by Intel, NVIDIA, Hudson River Trading, Qualcomm, Oculus, Amazon, IBM Research, Osmo, Occipital, and one generous 15-418 alumnus.