Jialong Li

Jialong Li

Auto-generated profile, based on publicly available information

Jialong Li

Shenzhen University of Advanced Technology, Shenzhen, China

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
22 July 2019 in IEEE/ACM Transactions on Networking
Verified
Journal Article
Closed access
ZZZhizhen Zhong
Zhizhen Zhong
NHNan Hua
Nan Hua
MTMassimo Tornatore
Massimo Tornatore
JLJialong Li
Jialong Li
YLYanhe Li
Yanhe Li
BMBiswanath Mukherjee
Biswanath Mukherjee
Transient traffic spikes are becoming a crucial challenge for network operators from both user-experience and network-maintenance perspectives. Different from long-term traffic growth, the bursty nature of short-term traffic fluctuations makes it difficult to be provisioned effectively. Luckily, next-generation elastic optical networks (EONs) provide an economical way to deal with such short-term traffic fluctuations. In this paper, we go beyond conventional network reconfiguration approaches by proposing the novel lightpath-splitting scheme in EONs. In lightpath splitting, we introduce the concept of SplitPoints to describe how lightpath splitting is performed. Lightpaths traversing multiple nodes in the optical layer can be split into shorter ones by SplitPoints to serve more traffic demands by raising signal modulation levels of lightpaths accordingly. We formulate the problem into a mathematical optimization model and linearize it into an integer linear program (ILP). We solve the optimization model on a small network instance and design scalable heuristic algorithms based on greedy and simulated annealing approaches. Numerical results show the tradeoff between throughput gain and negative impacts like traffic interruptions. Especially, by selecting SplitPoints wisely, operators can achieve almost twice as much throughput as conventional schemes without lightpath splitting.

1 May 2019 in IEEE Access
Verified
Journal Article
Open access
ZZZelin Zheng
Zelin Zheng
NHNan Hua
Nan Hua
ZZZhizhen Zhong
Zhizhen Zhong
JLJialong Li
Jialong Li
YLYanhe Li
Yanhe Li
XZXiaoping Zheng
Xiaoping Zheng
In future optical satellite networks with various service requirements, the bandwidth of a single traffic request occupies part of an inter-satellite link (ISL) channel capacity, thus leading to a greater demand for flexible resource allocation. The switching scheme is the most important determinant for flexible resource allocation in optical satellite networks, whereas the widely used optical switching techniques, including wavelength switching (WS) and electronic packet switching (EPS), have their own drawbacks such as bandwidth underutilization and high power consumption. In this paper, we utilize optical time slice switching (OTSS) to provide enormous transparent fine-grained connections for various traffic requests. To implement OTSS into low earth orbit (LEO) satellite networks, where the lengths of inter-orbit ISLs vary along time, we propose a distance-varying routing and time slice allocation (DV-RTSA) scheme. Simulation results demonstrate that the OTSS-based DV-RTSA scheme not only utilizes less active transponders when compared with the traditional schemes, but also achieves a significant throughput increase over the WSbased scheme and reaches the level of the EPS-based scheme exempting from its defects.

4 August 2024
Verified
Conference Paper
Open access
JLJialong Li
Jialong Li
HGHaotian Gong
Haotian Gong
FMFederico De Marchi
Federico De Marchi
AGAoyu Gong
Aoyu Gong
YLYiming Lei
Yiming Lei
YXYiting Xia
Yiting Xia
Reconfigurable data center networks (RDCNs) are arising as a promising data center network (DCN) design in the post-Moore's law era. However, the constantly reconfigured network topology in RDCNs invalidates the assumption of using hop count as the cost metric for routing, e.g., the status quo Equal-Cost Multi-Path routing (ECMP) in traditional DCNs. Unfortunately, existing routing solutions in RDCNs stick to the old assumption and deliver suboptimal performance either high in latency or low in bandwidth efficiency. In this paper, we redefine the cost metric for RDCN routing with uniform cost to unify the effects of topology disruption and hop count on latency and bandwidth efficiency. We propose Uniform-Cost Multi-Path routing (UCMP), an ECMP equivalent for RDCNs, where minimizing uniform cost leads flows of various sizes to the right balance between latency and bandwidth efficiency. Our simulation shows that UCMP achieves 53% to 98% lower flow completion time (FCT) and 1.55× bandwidth efficiency compared to the state-of-the-art RDCN routing strategy, and our testbed implementation demonstrates sustainable switch resource usage of UCMP as RDCNs scale.

21 November 2019 in Journal of Optical Communications and Networking
Verified
Journal Article
Closed access
PDF file
JLJialong Li
Jialong Li
NHNan Hua
Nan Hua
ZZZhizhen Zhong
Zhizhen Zhong
YYYufang Yu
Yufang Yu
XZXiaoping Zheng
Xiaoping Zheng
BZBingkun Zhou
Bingkun Zhou
The “pay-as-you-grow” cloud computing model has become popular for today’s enterprises. Cloud computing not only frees end users from complex operations, but also allows higher resource utilization, lower investment, and increased energy efficiency. However, with some emerging technologies, cloud computing is unable to meet the required latency level, especially for delay-sensitive services such as 5G communications, live streaming, and online gaming. In this context, edge computing is regarded as a promising technology that can provide low-latency connections in the near future. Unlike cloud computing, which provides service in a centralized mode, edge computing deploys micro data centers (MDCs) at the edge of the network to provide rapid-response service. However, due to the limited computing and storage capacity of a single MDC, a user may not be able to access the resources from the closest MDC during peak traffic periods. Under such circumstances, the user is served through another MDC or a remote cloud data center. The data are processed in an optical line terminal and then transmitted via the metro network, which significantly increases the latency. In this study, we introduce a flexible low-latency metro-access converged network architecture based on optical time slice switching (OTSS) to address the latency problem. By leveraging the transparent connections of the OTSS in this novel architecture, data can be transmitted through the MDCs without requiring extra processing time. The simulation results demonstrate that our proposed architecture can provide lower-latency connections under a range of conditions, with a negligible decrease in network throughput, compared with an existing representative architecture. Additionally, we conducted experiments to validate the feasibility of our approach.

18 December 2025 in IEEE Transactions on Networking
Verified
Journal Article
Closed access
JLJialong Li
Jialong Li
STShreyansh Tripathi
Shreyansh Tripathi
LRLakshay Rastogi
Lakshay Rastogi
YLYiming Lei
Yiming Lei
RPRui Pan
Rui Pan
YXYiting Xia
Yiting Xia
As machine learning models scale in size and complexity, their computational requirements become a significant barrier. Mixture-of-Experts (MoE) models alleviate this issue by selectively activating relevant experts. Despite this, MoE models are hindered by high communication overhead from all-to-all operations, low GPU utilization, and complications from heterogeneous GPU environments. This paper presents Comet, which optimizes both model deployment and all-to-all communication scheduling to address these challenges in MoE inference. Comet achieves minimal communication times by strategically ordering token transmissions in all-to-all communications. It improves GPU utilization by colocating experts from different models on the same device, avoiding the limitations of all-to-all communication. We analyze Comet’s optimization strategies theoretically across four common GPU cluster settings: exclusive vs. colocated models on GPUs, and homogeneous vs. heterogeneous GPUs. Comet provides optimal solutions for three cases, and for the remaining NP-hard scenario, it offers a polynomial-time sub-optimal solution with only a 1.09× degradation from the optimal, as shown in the simulation results. Comet is the first approach to minimize MoE inference time via optimal model deployment and communication scheduling across various scenarios. Evaluations demonstrate that Comet significantly accelerates inference, achieving speedups of up to 2.63× in homogeneous clusters and 2.91× in heterogeneous environments. Moreover, Comet enhances GPU utilization by up to 2.38× compared to existing methods.

25 May 2017 in IEEE Communications Letters
Verified
Journal Article
Closed access
JLJialong Li
Jialong Li
ZZZhizhen Zhong
Zhizhen Zhong
NHNan Hua
Nan Hua
XZXiaoping Zheng
Xiaoping Zheng
BZBingkun Zhou
Bingkun Zhou
Energy-efficient time- and wavelength-division multiplexed passive optical network has been intensely investigated. However, conventional schemes aimed at energy efficiency may bring about repeated power-state transitions between sleep mode and active mode, resulting in periodic device-temperature cycling and frequent wavelength reassignment, which will greatly deteriorate network quality of service. These side effects become non-trivial when traffic fluctuates sharply as expected in the future. In this letter, we propose a wavelength-postponed-switching-off (WPS) strategy to reduce power-state transitions by means of postponing the power-off of redundant wavelengths. Redundant wavelengths represent wavelengths that are supposed to be powered off by conventional schemes when traffic declines. Illustrative results show that, the WPS strategy can provide a balanced performance of both energy efficiency and device lifetime, while decreasing service interruption caused by turning ON or OFF the equipment.