Yuebin Bai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yuebin Bai has a total of 77 co-authors. Here are the top co-authors by number of shared publications:
- Ming Liu
University of Wisconsin-Madison, Madison, Wisconsin, USA
4 shared publications Weitao Wang
Beihang Univ, Sch Comp Sci & Engn, Beijing, Peoples R China
4 shared publications- Chao Yu
School of Computer Science and Engineering, Beihang University, Beijing, Beijing China
3 shared publications Kun Cheng
3 shared publications- Kun Cheng
State Key Laboratory of Software Development Environment, Beihang University, Beijing, China
3 shared publications - Xu Shao
Institute for Infocomm Research
3 shared publications
Latest Publications
- Open Access
- 22 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 2141-2158
Understanding and Optimizing Database Pushdown on Disaggregated Storage
- 22 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 2141-2158
Open AccessUnderstanding and Optimizing Database Pushdown on Disaggregated Storage
- Open Access
- 27 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 758-779
Understanding and Profiling CXL.mem Using PathFinder
- 27 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 758-779
Open AccessUnderstanding and Profiling CXL.mem Using PathFinder
- 9 July 2024
- Published by Elsevier in Neural Networks
- Vol. 179, 106522
Probability graph complementation contrastive learning
- 9 July 2024
- Published by Elsevier in Neural Networks
- Vol. 179, 106522
Probability graph complementation contrastive learning
- Open Access
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 916-929
LogNIC: A High-Level Performance Model for SmartNICs
SmartNICs have become an indispensable communication fabric and computing substrate in today’s data centers and enterprise clusters, providing in-network computing capabilities for traversed packets and benefiting a range of applications across the system stack. Building an efficient SmartNIC-assisted solution is generally non-trivial and tedious as it requires programmers to understand the SmartNIC architecture, refactor application logic to match the device’s capabilities and limitations, and correlate an application execution with traffic characteristics. A high-level SmartNIC performance model can decouple the underlying SmartNIC hardware device from its offloaded software implementations and execution contexts, thereby drastically simplifying and facilitating the development process. However, prior architectural models can hardly be applied due to their limited capabilities in dissecting the SmartNIC-offloaded program’s complexity, capturing the nondeterministic overlapping between computation and I/O, and perceiving diverse traffic profiles. This paper presents the LogNIC model that systematically analyzes the performance characteristics of a SmartNIC-offloaded program. Unlike conventional execution flow-based modeling, LogNIC employs a packet-centric approach that examines SmartNIC execution based on how packets traverse heterogeneous computing domains, on-/off-chip interconnects, and memory subsystems. It abstracts away the low-level device details, represents a deployed program as an execution graph, retains a handful of configurable parameters, and generates latency/throughput estimation for a given traffic profile. It further exposes a couple of extensions to handle multi-tenancy, traffic interleaving, and accelerator peculiarity. We demonstrate the LogNIC model’s capabilities using both commodity SmartNICs and an academic prototype under five application scenarios. Our evaluations show that LogNIC can estimate performance bounds, explore software optimization strategies, and provide guidelines for new hardware designs.
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 916-929
Open AccessLogNIC: A High-Level Performance Model for SmartNICs
SmartNICs have become an indispensable communication fabric and computing substrate in today’s data centers and enterprise clusters, providing in-network computing capabilities for traversed packets and benefiting a range of applications across the system stack. Building an efficient SmartNIC-assisted solution is generally non-trivial and tedious as it requires programmers to understand the SmartNIC architecture, refactor application logic to match the device’s capabilities and limitations, and correlate an application execution with traffic characteristics. A high-level SmartNIC performance model can decouple the underlying SmartNIC hardware device from its offloaded software implementations and execution contexts, thereby drastically simplifying and facilitating the development process. However, prior architectural models can hardly be applied due to their limited capabilities in dissecting the SmartNIC-offloaded program’s complexity, capturing the nondeterministic overlapping between computation and I/O, and perceiving diverse traffic profiles. This paper presents the LogNIC model that systematically analyzes the performance characteristics of a SmartNIC-offloaded program. Unlike conventional execution flow-based modeling, LogNIC employs a packet-centric approach that examines SmartNIC execution based on how packets traverse heterogeneous computing domains, on-/off-chip interconnects, and memory subsystems. It abstracts away the low-level device details, represents a deployed program as an execution graph, retains a handful of configurable parameters, and generates latency/throughput estimation for a given traffic profile. It further exposes a couple of extensions to handle multi-tenancy, traffic interleaving, and accelerator peculiarity. We demonstrate the LogNIC model’s capabilities using both commodity SmartNICs and an academic prototype under five application scenarios. Our evaluations show that LogNIC can estimate performance bounds, explore software optimization strategies, and provide guidelines for new hardware designs.
- Open Access
- 10 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 1012-1027
LEED: A Low-Power, Fast Persistent Key-Value Store on SmartNIC JBOFs
The recent emergence of low-power high-throughput programmable storage platforms-SmartNIC JBOF (just-a-bunch-of-flash)-motivates us to rethink the cluster architecture and system stack for energy-efficient large-scale data-intensive workloads. Unlike conventional systems that use an array of server JBOFs or embedded storage nodes, the introduction of SmartNIC JBOFs has drastically changed the cluster compute, memory, and I/O configurations. Such an extremely imbalanced architecture makes prior system design philosophies and techniques either ineffective or invalid. This paper presents LEED, a distributed, replicated, and persistent key-value store over an array of SmartNIC JBOFs. Our key ideas to tackle the unique challenges induced by a SmartNIC JBOF are: trading excessive I/O bandwidth for scarce SmartNIC core computing cycles and memory capacity; making scheduling decisions as early as possible to streamline the request execution flow. LEED systematically revamps the software stack and proposes techniques across per-SSD, intra-JBOF, and inter-JBOF levels. Our prototyped system based on Broadcom Stingray outperforms existing solutions that use beefy server JBOFs and wimpy embedded storage nodes by 4.2×/3.8× and 17.5×/19.1× in terms of requests per Joule for 256B/1KB key-value objects.
- 10 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 1012-1027
Open AccessLEED: A Low-Power, Fast Persistent Key-Value Store on SmartNIC JBOFs
The recent emergence of low-power high-throughput programmable storage platforms-SmartNIC JBOF (just-a-bunch-of-flash)-motivates us to rethink the cluster architecture and system stack for energy-efficient large-scale data-intensive workloads. Unlike conventional systems that use an array of server JBOFs or embedded storage nodes, the introduction of SmartNIC JBOFs has drastically changed the cluster compute, memory, and I/O configurations. Such an extremely imbalanced architecture makes prior system design philosophies and techniques either ineffective or invalid. This paper presents LEED, a distributed, replicated, and persistent key-value store over an array of SmartNIC JBOFs. Our key ideas to tackle the unique challenges induced by a SmartNIC JBOF are: trading excessive I/O bandwidth for scarce SmartNIC core computing cycles and memory capacity; making scheduling decisions as early as possible to streamline the request execution flow. LEED systematically revamps the software stack and proposes techniques across per-SSD, intra-JBOF, and inter-JBOF levels. Our prototyped system based on Broadcom Stingray outperforms existing solutions that use beefy server JBOFs and wimpy embedded storage nodes by 4.2×/3.8× and 17.5×/19.1× in terms of requests per Joule for 256B/1KB key-value objects.
- 16 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 34 (7) , 2091-2106
CD-MSA: Cooperative and Deadline-Aware Scheduling for Efficient Multi-Tenancy on DNN Accelerators
With DNN turning into the backbone of AI cloud services and propelling the emergence of INFerence-as-a-Service (INFaaS), DNN-specific accelerators have become the indispensable components of cloud inference systems. Due to the conservative “one-task-at-a-time” working mode and deadline blindness of those accelerators, implementing multi-tenancy that aims to improve the cost-effectiveness and meet SLA requirements is intractable. Recent studies including the temporal and spatial approaches, employ manifold scheduling mechanisms and sophisticated architecture innovations to address the challenge. However, these researches either still neglect the deadline awareness or render inevitable and expensive hardware overheads such as switches and storage. In this paper, we present Cooperative and Deadline-aware Multi-Systolic-Array scheduling (CD-MSA), a low-cost solution for the cloud inference that utilizes the real time mechanism and task-level parallelism to enable efficient multi-tenancy. Based on our preemptive multi-systolic-array accelerator architecture supporting the simultaneous task co-location, we first construct a fine-grained DNN execution model to lay the groundwork for the lightweight preemption. Second, we design a cooperative, deadline- and laxity-aware scheduler in conjunction with an efficient schedulability test method for better QoS guarantee without introducing additional hardware cost. Finally, to further promote the overall throughput, we propose dynamic task fusion , a software approach that fuses different tasks into the logically “multi-threading” tasks at runtime. We compare CD-MSA with several state-of-the-art researches across three multi-DNN workloads. The evaluation results show CD-MSA improves the latency-bounded throughput, SLA satisfaction rate and weighted system throughput by up to 62%, 63% and 27%, respectively.
- 16 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 34 (7) , 2091-2106
CD-MSA: Cooperative and Deadline-Aware Scheduling for Efficient Multi-Tenancy on DNN Accelerators
With DNN turning into the backbone of AI cloud services and propelling the emergence of INFerence-as-a-Service (INFaaS), DNN-specific accelerators have become the indispensable components of cloud inference systems. Due to the conservative “one-task-at-a-time” working mode and deadline blindness of those accelerators, implementing multi-tenancy that aims to improve the cost-effectiveness and meet SLA requirements is intractable. Recent studies including the temporal and spatial approaches, employ manifold scheduling mechanisms and sophisticated architecture innovations to address the challenge. However, these researches either still neglect the deadline awareness or render inevitable and expensive hardware overheads such as switches and storage. In this paper, we present Cooperative and Deadline-aware Multi-Systolic-Array scheduling (CD-MSA), a low-cost solution for the cloud inference that utilizes the real time mechanism and task-level parallelism to enable efficient multi-tenancy. Based on our preemptive multi-systolic-array accelerator architecture supporting the simultaneous task co-location, we first construct a fine-grained DNN execution model to lay the groundwork for the lightweight preemption. Second, we design a cooperative, deadline- and laxity-aware scheduler in conjunction with an efficient schedulability test method for better QoS guarantee without introducing additional hardware cost. Finally, to further promote the overall throughput, we propose dynamic task fusion , a software approach that fuses different tasks into the logically “multi-threading” tasks at runtime. We compare CD-MSA with several state-of-the-art researches across three multi-DNN workloads. The evaluation results show CD-MSA improves the latency-bounded throughput, SLA satisfaction rate and weighted system throughput by up to 62%, 63% and 27%, respectively.
- Open Access
PiPAD: Pipelined and Parallel Dynamic GNN Training on GPUs
Dynamic Graph Neural Networks (DGNNs) have been broadly applied in various real-life applications, such as link prediction and pandemic forecast, to capture both static structural information and temporal characteristics from dynamic graphs. Combining both time-dependent and -independent components, DGNNs manifest substantial parallel computation and data reuse potentials, but suffer from severe memory access inefficiency and data transfer overhead under the canonical one-graph-at-a-time training pattern. To tackle the challenges, we propose PiPAD, a $\underline{\textbf{Pi}}pelined$ and $\underline{\textbf{PA}}rallel$ $\underline{\textbf{D}}GNN$ training framework for the end-to-end performance optimization on GPUs. From both the algorithm and runtime level, PiPAD holistically reconstructs the overall training paradigm from the data organization to computation manner. Capable of processing multiple graph snapshots in parallel, PiPAD eliminates the unnecessary data transmission and alleviates memory access inefficiency to improve the overall performance. Our evaluation across various datasets shows PiPAD achieves $1.22\times$-$9.57\times$ speedup over the state-of-the-art DGNN frameworks on three representative models.Open AccessPiPAD: Pipelined and Parallel Dynamic GNN Training on GPUs
Dynamic Graph Neural Networks (DGNNs) have been broadly applied in various real-life applications, such as link prediction and pandemic forecast, to capture both static structural information and temporal characteristics from dynamic graphs. Combining both time-dependent and -independent components, DGNNs manifest substantial parallel computation and data reuse potentials, but suffer from severe memory access inefficiency and data transfer overhead under the canonical one-graph-at-a-time training pattern. To tackle the challenges, we propose PiPAD, a $\underline{\textbf{Pi}}pelined$ and $\underline{\textbf{PA}}rallel$ $\underline{\textbf{D}}GNN$ training framework for the end-to-end performance optimization on GPUs. From both the algorithm and runtime level, PiPAD holistically reconstructs the overall training paradigm from the data organization to computation manner. Capable of processing multiple graph snapshots in parallel, PiPAD eliminates the unnecessary data transmission and alleviates memory access inefficiency to improve the overall performance. Our evaluation across various datasets shows PiPAD achieves $1.22\times$-$9.57\times$ speedup over the state-of-the-art DGNN frameworks on three representative models. - 13 February 2021
- Published by Springer Nature in Wireless Personal Communications
- Vol. 118 (1) , 575-598
DTN-Balance: A Forwarding-Capacity and Forwarding-Queue Aware Routing for Self-organizing DTNs
- 13 February 2021
- Published by Springer Nature in Wireless Personal Communications
- Vol. 118 (1) , 575-598
DTN-Balance: A Forwarding-Capacity and Forwarding-Queue Aware Routing for Self-organizing DTNs
- 1 January 2021
- Published by Desalination Publications in Desalination and Water Treatment
- Vol. 213, 84-90
Water quality assessment based on interval-valued data cluster analysis
- 1 January 2021
- Published by Desalination Publications in Desalination and Water Treatment
- Vol. 213, 84-90
Water quality assessment based on interval-valued data cluster analysis
- 23 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 70 (11) , 1804-1816
MIPSGPU: Minimizing Pipeline Stalls for GPUs With Non-Blocking Execution
Improving the latency hiding ability is important for GPU performance. Although existing works, which mainly target on either improving thread level parallelism or optimizing memory hierarchy, are effective at improving GPUs’ latency hiding ability, warps are still blocked after executing long latency operations, reducing the number of schedulable warps. This article revisits the recently proposed non-blocking execution for GPUs to improve the latency hiding ability of GPUs. With non-blocking execution, instructions from warps blocked by long latency operations can be pre-executed to make full use of GPU resources. However, we find that the state-of-the-art non-blocking GPU architecture gains limited performance improvement. Through in-depth analysis, we observe that the poor performance is largely due to inefficient pre-execution state management, duplicate instruction extraction, frequent early eviction and severe resource congestion. To make non-blocking execution actually useful for GPUs and minimize hardware overheads, we carefully redesign the non-blocking architecture for GPUs based on our analysis and propose MIPSGPU . Our evaluations show that MIPSGPU , relative to the state-of-the-art non-blocking GPU architecture, improves performance of memory intensive applications by 19.05 percent, and reduces memory to SM traffics by 14 percent.
- 23 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 70 (11) , 1804-1816
MIPSGPU: Minimizing Pipeline Stalls for GPUs With Non-Blocking Execution
Improving the latency hiding ability is important for GPU performance. Although existing works, which mainly target on either improving thread level parallelism or optimizing memory hierarchy, are effective at improving GPUs’ latency hiding ability, warps are still blocked after executing long latency operations, reducing the number of schedulable warps. This article revisits the recently proposed non-blocking execution for GPUs to improve the latency hiding ability of GPUs. With non-blocking execution, instructions from warps blocked by long latency operations can be pre-executed to make full use of GPU resources. However, we find that the state-of-the-art non-blocking GPU architecture gains limited performance improvement. Through in-depth analysis, we observe that the poor performance is largely due to inefficient pre-execution state management, duplicate instruction extraction, frequent early eviction and severe resource congestion. To make non-blocking execution actually useful for GPUs and minimize hardware overheads, we carefully redesign the non-blocking architecture for GPUs based on our analysis and propose MIPSGPU . Our evaluations show that MIPSGPU , relative to the state-of-the-art non-blocking GPU architecture, improves performance of memory intensive applications by 19.05 percent, and reduces memory to SM traffics by 14 percent.