Junyi (Kyle) Shu

151 N. Zhongguancun Ave

Yanyuan Mansion, Rm 818

Beijing, China

I am a third-year Ph.D. candidate at Peking University, co-advised by Prof. Hong Mei and Prof. Xin Jin. My research interests lie in building disaggregated storage systems atop modern network/storage hardware (e.g., RDMA NIC, DPU & NVMe SSD).

I graduated from University of California at Berkeley with a Bachelor’s degree in Applied Mathematics and Computer Science in 2013. I joined Amazon Web Services after that, where I built distributed systems for account, product, and subscription management. I have also worked for companies in telecommunication sector, such as China Tower and China Telecom Bestpay.

I am a foodie and enjoy making new dishes. Japanese, Italian, Vietnamese and Mediterranean are among my favorites. I am also a wine lover. I passed WSET Level 2 with distinction and may pursue Level 3 someday. I like Gewürztraminer, Sauvignon Blanc, Riesling, and Pinot Noir. These days, I’m really into red wine from South France and Italy.

news

Mar 22, 2024	Burstable Cloud Block Storage with Data Processing Units is accepted by OSDI ‘24.
Feb 6, 2023	Disaggegated RAID Storage in Modern Datacenters is accepted by ASPLOS ‘23.
Dec 6, 2022	Successfully passed the Ph.D. qualifying examination. Many thanks to Prof. Tao Xie, Prof. Minghui Zhou, Prof. Xuanzhe Liu, Prof. Xin Jin, and Prof. Xin Zhang.

selected publications

ASPLOS

Disaggregated RAID Storage in Modern Datacenters

Junyi Shu, Ruidong Zhu, Yun Ma, Gang Huang, Hong Mei, Xuanzhe Liu, and Xin Jin

In Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3 2023

Abs PDF Code Poster Slides

RAID (Redundant Array of Independent Disks) has been widely adopted for decades, as it provides enhanced throughput and redundancy beyond what a single disk can offer. Today, enabled by fast datacenter networks, accessing remote block devices with acceptable overhead (i.e. disaggregated storage) becomes a reality (e.g., for serverless applications). Combining RAID with remote storage can provide the same benefits while creating better fault tolerance and flexibility than its monolithic counterparts. The key challenge of disaggregated RAID is to handle extra network traffic generated by RAID, which can consume a vast amount of NIC bandwidth. We present dRAID, a disaggregated RAID system that achieves near-optimal read and write throughput. dRAID exploits peer-to-peer disaggregated data access to reduce bandwidth consumption in both normal and degraded states. It employs non-blocking multi-stage writes to maximize inter-node parallelism, and applies pipelined I/O processing to maximize inter-device parallelism. We introduce bandwidth-aware reconstruction for better load balancing. We show that dRAID provides up to 3x bandwidth improvement. The results on a lightweight object store show that dRAID brings 1.5x-2.35x throughput improvement on various workloads.