Yanfeng Zhang

Dr. Yanfeng Zhang


Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 December 2020 in IEEE Access
Verified
Journal Article
Open access
GFGuo Fu
Guo Fu
YZYanfeng Zhang
Yanfeng Zhang
GYGe Yu
Ge Yu
The production and real-time usage of streaming data bring new challenges for data systems due to huge volume of streaming data and quick response request of applications. Message queuing systems that offer high throughput and low latency play an important role in today’s big streaming data processing. There are several popular message queuing systems in production usage and also many in-lab message queuing systems in academia. These systems with different design philosopies have different characteristics. It is non-trivial for a non-expert to choose a suitable system to meet his specific requirement. With this premise, our primary contribution is to provide the community with a fair comparison among message queuing systems, using a standardized comparison metric and reproducible experimental environment. Five typical message queuing systems (including Kafka, RabbitMQ, RocketMQ, ActiveMQ and Pulsar) are evaluated qualitatively (in analysis) and quantitatively (in experimental results). This article also highlights the distinct features of each system and summarizes the best-suited use cases of each system. The fair comparison and the insight analysis provided in this article can help users choose the best-suited message queuing systems.

18 December 2024 in Proceedings of the ACM on Management of Data
Verified
Journal Article
Open access
SYSong Yu
Song Yu
SGShufeng Gong
Shufeng Gong
QTQian Tao
Qian Tao
SSSijie Shen
Sijie Shen
YZYanfeng Zhang
Yanfeng Zhang
JZJingren Zhou
Jingren Zhou
The growing volume of graph data may exhaust the main memory. It is crucial to design a disk-based graph storage system to ingest updates and analyze graphs efficiently. However, existing dynamic graph storage systems suffer from read or write amplification and face the challenge of optimizing both read and write performance simultaneously. To address this challenge, we propose LSMGraph, a novel dynamic graph storage system that combines the write-friendly LSM-tree and the read-friendly CSR. It leverages the multi-level structure of LSM-trees to optimize write performance while utilizing the compact CSR structures embedded in the LSM-trees to boost read performance. LSMGraph uses a new memory structure, MemGraph, to efficiently cache graph updates and uses a multi-level index to speed up reads within the multi-level structure. Furthermore, LSMGraph incorporates a vertex-grained version control mechanism to mitigate the impact of LSM-tree compaction on read performance and ensure the correctness of concurrent read and write operations. Our evaluation shows that LSMGraph significantly outperforms state-of-the-art (graph) storage systems on both graph update and graph analytical workloads.

26 May 2023 in Proceedings of the ACM on Management of Data
Verified
Journal Article
Open access
WZWeixing Zhou
Weixing Zhou
QPQi Peng
Qi Peng
ZZZijie Zhang
Zijie Zhang
YZYanfeng Zhang
Yanfeng Zhang
YRYang Ren
Yang Ren
GYGe Yu
Ge Yu
Multinational enterprises conduct global business that has a demand for geo-distributed transactional databases. Existing state-of-the-art databases adopt a sharded master-follower replication architecture. However, the single-master serving mode incurs massive cross-region writes from clients, and the sharded architecture requires multiple round-trip acknowledgments (e.g., 2PC) to ensure atomicity for cross-shard transactions. These limitations drive us to seek yet another design choice. In this paper, we propose a strongly consistent OLTP database GeoGauss with full replica multi-master architecture. To efficiently merge the updates from different master nodes, we propose a multi-master OCC that unifies data replication and concurrent transaction processing. By leveraging an epoch-based delta state merge rule and the optimistic asynchronous execution, GeoGauss ensures strong consistency with light-coordinated protocol and allows more concurrency with weak isolation, which are sufficient to meet our needs. Our geo-distributed experimental results show that GeoGauss achieves 7.06X higher throughput and 17.41X lower latency than the state-of-the-art geo-distributed database CockroachDB on the TPC-C benchmark.

29 November 2022 in Future Generation Computer Systems
Verified
Journal Article
Closed access
ZLZhuojin Liu
Zhuojin Liu
SGShufeng Gong
Shufeng Gong
YSYuxuan Su
Yuxuan Su
CWChangyi Wan
Changyi Wan
YZYanfeng Zhang
Yanfeng Zhang
GYGe Yu
Ge Yu

9 November 2022 in World Wide Web
Verified
Journal Article
Closed access
PDF file
GWGang Wang
Gang Wang
QPQi Peng
Qi Peng
YZYanfeng Zhang
Yanfeng Zhang
MZMingyang Zhang
Mingyang Zhang

8 December 2023 in Proceedings of the ACM on Management of Data
Verified
Journal Article
Open access
ZSZhen Song
Zhen Song
YGYu Gu
Yu Gu
TLTIANYI LI
TIANYI LI
QSQing Sun
Qing Sun
YZYanfeng Zhang
Yanfeng Zhang
GYGe Yu
Ge Yu
Distributed computing is promising to enable large-scale graph neural network (GNN) model training. However, care is needed to avoid excessive computational and communication overheads. Sampling is promising in terms of enabling scalability, and sampling techniques have been proposed to reduce training costs. However, online sampling introduces large overheads, and while offline sampling that is done only once can eliminate such overheads, it instead introduces information loss and accuracy degradation. Thus, existing sampling techniques are unable to improve simultaneously both efficiency and accuracy, particularly at low sampling rates. We develop a distributed system, ADGNN, for full-batch based GNN training that adopts a hybrid sampling architecture to enable a trade-off between efficiency and accuracy. Specifically, ADGNN employs sampling result reuse techniques to reduce the cost associated with sampling and thus improve training efficiency. To alleviate accuracy degradation, we introduce a new metric,Aggregation Difference (AD), that quantifies the gap between sampled and full neighbor set aggregation. We present so-called AD-Sampling that aims to minimize the Aggregation Difference with an adaptive sampling frequency tuner. Finally, ADGNN employs anAD -importance-based sampling technique for remote neighbors to further reduce communication costs. Experiments on five real datasets show that ADGNN is able to outperform the state-of-the-art by up to nearly 9 times in terms of efficiency, while achieving comparable accuracy to the non-sampling methods.