Haoran(Holland) Ma |

About Me

Hi, my name’s Haoran Ma. You can call me Holland. I’m currently a Research Scientist at ByteDance. I earned my PhD in Computer Science from University of California, Los Angeles in 2024. I was co-advised by Prof. Harry Xu and Prof. Miryung Kim during my PhD studies.

Education

University of California, Los Angeles

PhD in Computer Science

2019 - 2024

I am interested in pushing the boundaries of system capabilities within diverse datacenter environments, particularly for data analytics and machine learning applications. My research predominantly centers on innovating memory management strategies for next-generation datacenters. This involves exploring novel approaches in memory disaggregation and distributed shared memory, leveraging technologies such as RDMA (Remote Direct Memory Access). My work aims to optimize resource utilization and efficiency, addressing critical challenges in scalability and performance in complex datacenter architectures.

Tsinghua University

BEng in Computer Science

2015 - 2019

I took part in the National Olympiad in Informatics in 2014 and I won the gold medal, which granted me admission to Tsinghua University for studying computer science in advance. During my studies at Tsinghua, I did research on Web API Testing, advised by Prof. Xiaoying Bai. I graduated summa cum laude from Tsinghua in 2019.

University of Oxford

Visiting Student in New College

2017 - 2018

I was selected into Yinghua Scholars Program at Tsinghua, which provided me the opportunity to go to the University of Oxford as a visiting student for one year with a full scholarship. In Oxford, I earned a full GPA in every course I studied.

I really love the formal dinner at New College.

Publications

1

Haoran Ma, Yifan Qiao, Shi Liu, Shan Yu, Yuanjiang Ni, Qingda Lu, Jiesheng Wu, Qingda Lu, Miryung Kim, Harry Xu.

DRust: Language-Guided Distributed Shared Memory with Fine Granularity, Full Transparency, and Ultra Efficiency

Accepted to 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI ‘24)

2

Lei Chen, Shi Liu (co-first author), Chenxi Wang, Haoran Ma, Yifan Qiao, Zhe Wang, Chenggang Wu, Youyou Lu, Xiaobing Feng, Huimin Cui, Shan Lu, Harry Xu

A Tale of Two Paths: Toward a Hybrid Data Plane for Efficient Far-Memory Applications

Accepted to 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI ‘24)

3

Yifan Qiao, Zhenyuan Ruan, Haoran Ma, Adam Belay, Miryung Kim, Harry Xu.

Harvesting Idle Memory for Application-managed Soft State with Midas

Accepted to 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI ‘24)

4

Chenxi Wang, Yifan Qiao (co-first author), Haoran Ma, Shi Liu, Yiying Zhang, Wenguang Chen, Ravi Netravali, Miryung Kim, Guoqing Harry Xu.

Canvas: Isolated and Adaptive Swapping for Multi-Applications on Remote Memory

20th USENIX Symposium on Networked Systems Design and Implementation (NSDI ‘23)

5

Chenxi Wang, Haoran Ma (co-first author), Shi Liu, Yifan Qiao, Jonathan Eyolfson, and Christian Navasca, UCLA; Shan Lu University of Chicago and Microsoft; Guoqing Harry Xu, UCLA.

MemLiner: Lining up Tracing and Application for a Far-Memory-Friendly Runtime

16th USENIX Symposium on Operating Systems Design and Implementation (OSDI ‘22). Awarded Jay Lepreau Best Paper

6

Haoran Ma, Shi Liu, Chenxi Wang, and Yifan Qiao, UCLA; Michael D. Bond, Ohio State University; Stephen M. Blackburn, Australian National University; Miryung Kim, and Guoqing Harry Xu, UCLA

Mako: A Low-Pause, High-Throughput Evacuating Collector for Memory-Disaggregated Datacenters

43rd ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI ‘22)

7

Chenxi Wang, Haoran Ma, Shi Liu, and Yuanqi Li, UCLA; Zhenyuan Ruan, MIT; Khanh Nguyen, Google and Texas A&M University; Michael D. Bond, Ohio State University; Ravi Netravali, Miryung Kim, and Guoqing Harry Xu, UCLA

Semeru: A Memory-Disaggregated Managed Runtime

14th USENIX Symposium on Operating Systems Design and Implementation (OSDI ‘20)