GG
Gregory R. Ganger
cs.DCcs.LGstat.MLcs.OSAlgorithmsComputer Graphicscs.AIcs.ARcs.DBcs.SE
On Valency
published · living versionsW_u7fa5rax·v1 · currentpublished
MLSys: The New Frontier of Machine Learning Systems
with Alexander Ratner, Dan Alistarh, Gustavo Alonso, David G. Andersen +64
1 version
Preprints & journals
13 papers in the corpus · 2014–2025Nonuniform-Tensor-Parallelism: Mitigating GPU failure impact for Scaled-up LLM Training2504.06095v1 · Daiyaan Arfeen, Dheevatsa Mudigere, Ankit More et al.2025 · 0 citationsarXiv
GraphPipe: Improving Performance and Scalability of DNN Training with Graph Pipeline Parallelism2406.17145v2 · Byungsoo Jeon, Mengdi Wu, Shiyi Cao et al.2024 · 10 citationsarXiv
PipeFill: Using GPUs During Bubbles in Pipeline-parallel LLM Training2410.07192v1 · Daiyaan Arfeen, Zhen Zhang, Xinwei Fu et al.2024 · 0 citationsarXiv
Pollux: Co-adaptive Cluster Scheduling for Goodput-Optimized Deep Learning2008.12260v2 · Aurick Qiao, Sang Keun Choe, Suhas Jayaram Subramanya et al.2020 · 50 citationsarXiv
PACEMAKER: Avoiding HeART attacks in storage clusters with disk-adaptive redundancy2103.08191v1 · Saurabh Kadekodi, Francisco Maturana, Suhas Jayaram Subramanya et al.2021 · 10 citations14th USENIX Symposium on Operating Systems Design and Implementation (OSDI), 2020, (pp. 369-385)
Vilamb: Low Overhead Asynchronous Redundancy for Direct Access NVM2004.09619v1 · Rajat Kateja, Andy Pavlo, Gregory R. Ganger2020 · 3 citationsarXiv
MLSys: The New Frontier of Machine Learning Systems1904.03257v3 · Alexander Ratner, Dan Alistarh, Gustavo Alonso et al.2019 · 20 citationsarXivon Valency
More Effective Distributed ML via a Stale Synchronous Parallel Parameter Server.25400488 · Ho, Qirong, Cipar, James, Cui, Henggang et al.2019 · 547 citationsAdvances in neural information processing systems. 2013;2013:1223-1231
Accelerating Deep Learning by Focusing on the Biggest Losers1910.00762v1 · Angela H. Jiang, Daniel L.-K. Wong, Giulio Zhou et al.2019 · 33 citationsarXiv
Tvarak: Software-managed hardware offload for DAX NVM storage redundancy1908.09922v1 · Rajat Kateja, Nathan Beckmann, Gregory R. Ganger2019 · 1 citationarXiv
PipeDream: Fast and Efficient Pipeline Parallel DNN Training1806.03377v1 · Aaron Harlap, Deepak Narayanan, Amar Phanishayee et al.2018 · 97 citationsarXiv
MLtuner: System Support for Automatic Machine Learning Tuning1803.07445v1 · Henggang Cui, Gregory R. Ganger, Phillip B. Gibbons2018 · 5 citationsarXiv
Visualizing request-flow comparison to aid performance diagnosis in distributed systems.24051813 · Sambasivan, Raja R, Shafer, Ilari, Mazurek, Michelle L et al.2014 · 18 citationsIEEE transactions on visualization and computer graphics. 2013;19(12):2466-75
Career total: 363 works. 13 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.