Project Akita

Project Akita builds the next generation high-performance, high-flexibility computer architecture simulation framework, with a special focus on developer experience.

Sub-Projects

Akita Simulator Engine

https://github.com/sarchlab/akita

The Akita Simulator serves as the foundational framework for constructing various simulators, offering libraries to streamline development. It includes essential models for caches and DRAM controllers, as well as a suite of tools designed for developer efficiency, including visualization, monitoring, and performance analysis.

ArchSim

ArchSim enables computer architecture simulation as a service. ArchSim provides a model zoo for computer architecture simulator models built with Akita and allows users to configure, execute, and analyze simulations in a single web-based platform.

MGPUSim

https://github.com/sarchlab/mgpusim

MGPUSim is a GPU simulator that is specialized for Multi-GPU simulation. It models AMD GPUs running OpenCL programs.

Zeonica

Zeonica is an on-going project that models Coarse-Grained Reconfigurable Array (CGRA) devices at a high level.

Events

The First Lightweight Workshop on Akita and MGPUSim (Akita '24)

Enabled Research

So far, there are 18 published papers that uses Akita-based simulators to validate design ideas or use Akita to build new simulators. Out of these 18 papers, 9 papers do not directly involve Dr. Sun. If we missed any other papers, please let us know.

Below is a full list of papers:

Yuan Feng, Seonjin Na, Hyesoon Kim, Hyeran Jeon. "Barre Chord: Efficient Virtual Memory Translation for Multi-Chip-Module GPUs " (2024)
Yueqi Wang, Bingyao Li, Aamer Jaleel, Jun Yang, Xulong Tang. "GRIT: Enhancing Multi-GPU Performance with Fine-Grained Dynamic Page Placement" (2024)
Bingyao Li, Yueqi Wang, Tianyu Wang, Lieven Eeckhout, Jun Yang, Aamer Jaleel, Xulong Tang. "Improving Multi-Instance GPU Efficiency via Sub-Entry Sharing TLB Design" (2024)
Seonjin Na, Jungwoo Kim, Sunho Lee, Jaehyuk Huh. "Supporting Secure Multi-GPU Computing with Dynamic and Batched Metadata Management" (2024)
Changxi Liu, Yifan Sun, and Trevor E. Carlson. "Photon: A Fine-grained Sampled Simulation Methodology for GPU Workloads." (2023).
Ying Li, Yifan Sun, and Adwait Jog. "Path Forward Beyond Simulators: Fast and Accurate GPU Execution Time Prediction for DNN Workloads." (2023).
Kaustubh Shivdikar, Yuhui Bao, Rashmi Agrawal, Michael Shen, Gilbert Jonatan, Evelio Mora, Alexander Ingare et al. "GME: GPU-based Microarchitectural Extensions to Accelerate Homomorphic Encryption." arXiv preprint arXiv:2309.11001 (2023).
Li, Bingyao, Yanan Guo, Yueqi Wang, Aamer Jaleel, Jun Yang, and Xulong Tang. "IDYLL: Enhancing Page Translation in Multi-GPUs via Light Weight PTE Invalidations." (2023).
Bingyao Li, Jieming Yin, Anup Holey, Youtao Zhang, Jun Yang, and Xulong Tang. "Trans-FW: Short Circuiting Page Table Walk in Multi-GPU Systems via Remote Forwarding." In 2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA), pp. 456-470. IEEE, 2023.
Belayneh, Leul, Haojie Ye, Kuan-Yu Chen, David Blaauw, Trevor Mudge, Ronald Dreslinski, and Nishil Talati. "Locality-Aware Optimizations for Improving Remote Memory Latency in Multi-GPU Systems." In Proceedings of the International Conference on Parallel Architectures and Compilation Techniques, pp. 304-316. 2022.
Yuihui Bao, Yifan Sun, Zlatan Feric, Michael Tian Shen, Micah Weston, José L. Abellán, Trinayan Baruah, John Kim, Ajay Joshi, and David Kaeli. "NaviSim: A Highly Accurate GPU Simulator for AMD RDNA GPUs." In Proceedings of the International Conference on Parallel Architectures and Compilation Techniques, pp. 333-345. 2022.
Michael Shen, Derek Rodriguez, and David Kaeli. "Evaluating the Impact of Branch Predictor Design on Spectre Attacks." (2022).
Shi Dong, Yifan Sun, Nicolas Bohm Agostini, Elmira Karimi, Daniel Lowell, Jing Zhou, José Cano, José L. Abellán, and David Kaeli. "Spartan: A sparsity-adaptive framework to accelerate deep neural network training on GPUs." IEEE Transactions on Parallel and Distributed Systems 32, no. 10 (2021): 2448-2463.
Bingyao Li, Jieming Yin, Youtao Zhang, and Xulong Tang. "Improving address translation in multi-gpus via sharing and spilling aware tlb design." In MICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture, pp. 1154-1168. 2021.
Griffin Knipe, and Derek Rodriguez. "RISC-V Microarchitecture Simulation State Enumeration." In Fifth Workshop on Computer Architecture Research with RISC-V (CARRV 2021). 2021.
Pengyu Wang, Jing Wang, Chao Li, Jianzong Wang, Haojin Zhu, and Minyi Guo. "Grus: Toward unified-memory-efficient high-performance graph processing on gpu." ACM Transactions on Architecture and Code Optimization (TACO) 18, no. 2 (2021): 1-25.
Trinayan Baruah, Yifan Sun, Ali Tolga Dinçer, Saiful A. Mojumder, José L. Abellán, Yash Ukidave, Ajay Joshi, Norman Rubin, John Kim, and David Kaeli. "Griffin: Hardware-software support for efficient page migration in multi-gpu systems." In 2020 IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 596-609. IEEE, 2020.
Trinayan, Baruah, Yifan Sun, Saiful A. Mojumder, José L. Abellán, Yash Ukidave, Ajay Joshi, Norman Rubin, John Kim, and David Kaeli. "Valkyrie: Leveraging inter-tlb locality to enhance gpu performance." In Proceedings of the ACM International Conference on Parallel Architectures and Compilation Techniques, pp. 455-466. 2020.
Saiful A. Mojumder, Yifan Sun, Leila Delshadtehrani, Yenai Ma, Trinayan Baruah, José L. Abellán, John Kim, David Kaeli, and Ajay Joshi. "Mgpu-tsm: A multi-gpu system with truly shared memory." arXiv preprint arXiv:2008.02300 (2020).
Saiful A. Mojumder, Yifan Sun, Leila Delshadtehrani, Yenai Ma, Trinayan Baruah, José L. Abellán, John Kim, David Kaeli, and Ajay Joshi. "Halcone: A hardware-level timestamp-based cache coherence scheme for multi-gpu systems." arXiv preprint arXiv:2007.04292 (2020).
Jinhui Wei, Jianzhuang Lu, Qi Yu, Chen Li, and Yunping Zhao. "Dynamic gmmu bypass for address translation in multi-gpu systems." In IFIP International Conference on Network and Parallel Computing, pp. 147-158. Cham: Springer International Publishing, 2020.
Chen Li, Yifan Sun, Lingling Jin, Lingjie Xu, Zheng Cao, Pengfei Fan, David Kaeli, Sheng Ma, Yang Guo, and Jun Yang. "Priority-based PCIe scheduling for multi-tenant multi-GPU systems." IEEE Computer Architecture Letters 18, no. 2 (2019): 157-160.