IEEE/ACM International Symposium on Quality of Service (IWQoS) 2020

Session Keynote-4: Keynote 4
Session 3A: Edge Computing Session 3B: Multimedia Session 3C: Security Session 3D: Network Optimization and Network Intelligence
Session Best-Paper: Best Paper and IWQoS 2021

Technical Sessions

Session 3A

Edge Computing

Conference
10:15 AM — 11:35 AM HKT
Local
Jun 16 Tue, 7:15 PM — 8:35 PM PDT

A Deep Reinforcement Learning Approach for Online Computation Offloading in Mobile Edge Computing

Yameng Zhang and Tong Liu (Shanghai University, China); Yanmin Zhu (Shanghai Jiao Tong University, China); Yuanyuan Yang (Stony Brook University, USA)

0
With the explosion of mobile smart devices, many computation intensive applications have emerged, such as interactive gaming and augmented reality. Mobile edge computing is put forward, as an extension of cloud computing, to meet the low-latency requirements of the applications. In this paper, we consider an edge computing system built in an ultra-dense network with numerous base stations, and heterogeneous computation tasks are successively generated on a smart device moving in the network. An optimal task offloading strategy, as well as optimal CPU frequency and transmit power scheduling, is desired by the device user, to minimize both task completion latency and energy consumption in a long term. However, due to the stochastic computation tasks and dynamic network conditions, the problem is particularly difficult to solve. Inspired by reinforcement learning, we transform the problem into a Markov decision process. Then, we propose an online offloading approach based on a double deep Q network, in which a specific neural network model is also provided to estimate the cumulative reward achieved by each action. We also conduct extensive simulations to compare the performance of our proposed approach with baselines.

Decode-and-Compare: An Efficient Verification Scheme for Coded Edge Computing

Mingjia Fu (Soochow University, China); Jin Wang (Soochow Univerisity & City University of Hong Kong, China); Jianping Wang (City University of Hong Kong, Hong Kong); Kejie Lu (University of Puerto Rico at Mayaguez, Puerto Rico); Admela Jukan (Technische Universität Carolo-Wilhelmina zu Braunschweig, Germany); Fei Gu (Soochow University, China)

1
Edge computing is a promising technology that can fulfill the requirements of latency-critical and computation-intensive applications. To further enhance the performance, coded edge computing has emerged because it can optimally utilize edge devices to speed up the computation. Nevertheless, to adopt coded edge computing, there are many challenges and one of the most important concerns is security. First, compared servers in a cloud computing scenario, the edge devices may not be reliable or trustworthy. Secondly, with coded computing, one modified intermediate result can screw up the final result. In this paper, we tackle a major security issue in coded edge computing: how to verify the correctness of results and identify attackers. Specifically, we propose an efficient verification scheme, namely Decode-and-Compare (DC), by leveraging both coding redundancy of edge devices and the properties of linear coding itself. To design the DC scheme, we conduct solid theoretical analysis to show the required coding redundancy, the expected number of decoding operations, the probability of successful verification, and the tradeoff among them. To evaluate the performance of DC, we conduct extensive simulation experiments and the results confirm that the DC scheme can outperform existing solutions, such as homomorphic encryption and computing locally at the user device.

Finedge: A Dynamic Cost-efficient Edge Resource Management Platform for NFV Network

Miao Li, Qixia Zhang and Fangming Liu (Huazhong University of Science and Technology, China)

0
With the evolution of network function virtualization (NFV) and edge computing, software-based network functions (NFs) can be deployed on closer-to-end-user edge servers to support a broad range of new services with high bandwidth and low latency. However, due to the resource limitation, strict QoS requirements and real-time flow fluctuations in edge network, existing cloud-based resource management strategy in NFV platforms is inefficient to be applied to the edge. Thus, we propose Finedge, a dynamic, fine-grained and cost-efficient edge resource management platform for NFV network. First, we conduct empirical experiments to find out the effect of NFs' resource allocation and their flow-level characteristics on performance. Then, by jointly considering these factors and QoS requirements (e.g., latency and packet loss rate), Finedge can automatically assign the most suitable CPU core and tune the most cost-efficient CPU quota to each NF. Finedge is also implemented with some key strategies including real-time flow monitoring, elastic resource scaling up and down, and also flexible NF migration among cores. Through extensive evaluations, we validate that Finedge can efficiently handle heterogeneous flows with the lowest CPU quota and the highest SLA satisfaction rate as compared with the default OS scheduler and other state-of-the-art resource management schemes.

Incentive Assignment in PoW and PoS Hybrid Blockchain in Pervasive Edge Environments

Yaodong Huang, Yiming Zeng, Fan Ye and Yuanyuan Yang (Stony Brook University, USA)

0
Edge computing is becoming pervasive in our daily lives with emerging smart devices and the development of communication technology. Resource-rich smart devices and high-density supportive communication technology make data transactions prevalent over edge environments. To ensure such transactions are unmodifiable and undeniable, blockchain technology is introduced into edge environments. In this paper, we propose a hybrid blockchain system in edge environments to enhance the security for transactions and determine the incentive for miners. We propose a PoS-PoW hybrid blockchain system utilizing the heterogeneity of devices to adapt to the characteristic of edge environments. We raise the incentive assignment problem that gives the corresponding PoW miner when a new block generates. We further formulate it into a two-stage Stackelberg game. We propose an algorithm and prove that it can obtain the global optimal results for the incentive that the miner will receive for a new block. Numerical simulation results show that our proposed algorithm can give reasonable incentive to miners under different system parameters in edge blockchain systems.
Session Chair

Rong Zheng (McMaster)

Session 3B

Multimedia

Conference
1:10 PM — 2:30 PM HKT
Local
Jun 16 Tue, 10:10 PM — 11:30 PM PDT

Delay-sensitive Computation Partitioning for Mobile Augmented Reality Applications

Chaokun Zhang (Tianjin University, China); Rong Zheng (McMaster University, Canada); Yong Cui (Tsinghua University, China); Chenhe Li (McMaster University, Canada); Jianping Wu (Tsinghua University, China)

3
Good user experiences in Mobile Augmented Reality (MAR) applications require timely processing and rendering of virtual objects on the user devices. Today's wearable AR devices are limited in computation, storage, and power sources. Edge computing, where edge devices are employed to offload part or all computation tasks allows acceleration of computation without incurring excessive network latency. In this paper, we use acyclic data flow graphs to model the computation and data flow in MAR applications and aim to minimize the makespan of processing input frames. Due to task dependencies and resource constraints, makespan minimization is proven to be NP-hard in general. We design DPA, a polynomial-time heuristic algorithm for this problem. For special data flow graphs including chain or star, the algorithm can provide optimal solutions or solutions with a constant approximation ratio. A prototype is implemented and evaluations under realistic profiles demonstrate the effectiveness of DPA.

Modeling and Analyzing Live Streaming Performance

Tong Zhang (Nanjing University of Aeronautics and Astronautics, China); Fengyuan Ren and Bo Wang (Tsinghua University, China)

1
Today, live streaming is gaining a rapid growth in use, which refers to streaming the media content recorded and broadcast in real time. In live streaming, latency is of utmost importance since smaller latency means higher user engagement. HTTP adaptive streaming (HAS) is now the most popular live streaming technology, where the video client sends HTTP requests to server to download video segments. The bitrate adaptation (ABR) algorithm inside the client determines bitrate level for every segment. It is of great help for ABR algorithm to quantify the influence of different HAS factors on streaming performance. However, existing work mainly focuses on video on demand (VoD) streaming rather than live streaming. In this paper, we theoretically analyze live streaming performance. We first establish a queuing model to describe playout buffer evolution. Based on the model, we respectively characterize rebuffering probability, rebuffering count and streaming latency, and analyze the effects of chunk arrival rate, arrival interval fluctuation, startup threshold and video skipping on them. From analysis results, we propose insights and recommendations for bitrate adaptation in live streaming and design a simple heuristic ABR algorithm leveraging them. Extensive simulations verify the insights as well as effectiveness of the designed algorithm.

Generative Adversarial Networks-based Privacy-Preserving for 3D Reconstruction

Qinya Li (Shanghai Jiaotong University, China); Zhenzhe Zheng, Fan Wu and Guihai Chen (Shanghai Jiao Tong University, China)

0
A large-scale image collection is crucial to the success of 3D reconstruction. Crowdsourcing, as a new pattern, can be utilized to collect high-quality images in an efficient way. However, the sensitive information in images may be exposed during the image transmission process. The general privacy policies perhaps will cause critical information loss or the change of information, which may give rise to a decline in the performance of 3D reconstruction. Hence, how to achieve image
privacy-preserving while guaranteeing to reconstruct a complete 3D model is important and significant. In this paper, we propose PicPrivacy to address this problem, which consists of three parts. (1) Using a pre-trained deep convolution neural network to identify sensitive information and erase it from images. (2) Using a GAN-based image feature completion algorithm to repair blank regions and minimize the absolute information gap between generated images and raw ones. (3) Taking generated images as the input of 3D reconstruction and using a structure-
from-motion algorithm to reconstruct 3D models. Finally, we extensively evaluate the performance of PicPrivacy on real-world datasets. The results demonstrate that PicPrivacy not only achieves individual privacy-preserving but also can guarantee to create complete 3D models.

PQA-CNN: Towards Perceptual Quality Assured Single-Image Super-Resolution in Remote Sensing

Yang Zhang, Xiangyu Dong, Md Tahmid Rashid, Lanyu Shang, Jun Han, Daniel Zhang and Dong Wang (University of Notre Dame, USA)

1
Recent advances in remote sensing open up unprecedented opportunities to obtain a rich set of visual features of objects on the earth's surface. In this paper, we focus on a single-image super-resolution (SISR) problem in remote sensing, where the objective is to generate a reconstructed satellite image of high quality (i.e., a high spatial resolution) from a satellite image of relatively low quality. This problem is motivated by the lack of high quality satellite images in many remote sensing applications (e.g., due to the cost of high resolution sensors, communication bandwidth constraints, and historic hardware limitations). Two important challenges exist in solving our problem: i) it is not a trivial task to reconstruct a satellite image of high quality that meets the human perceptual requirement from a single low quality image; ii) it is challenging to rigorously quantify the uncertainty of the results of an SISR scheme in the absence of ground truth data. To address the above challenges, we develop PQA-CNN, a perceptual quality-assured conventional neural network framework, to reconstruct a high quality satellite image from a low quality one by designing novel uncertainty-driven neural network architectures and integrating an uncertainty quantification model with the framework. We evaluate PQA-CNN on a real-world remote sensing application on land usage classifications. The results show that PQA-CNN significantly outperforms the state-of-the-art super-resolution baselines in terms of accurately reconstructing high-resolution satellite images under various evaluation scenarios.
Session Chair

Panlong Yang (USTC)

Session 3C

Security

Conference
2:50 PM — 4:10 PM HKT
Local
Jun 16 Tue, 11:50 PM — 1:10 AM PDT

Network-based Malware Detection with a Two-tier Architecture for Online Incremental Update

Anli Yan and Zhenxiang Chen (University of Jinan, China); Riccardo Spolaor (University of Oxford, United Kingdom (Great Britain)); Shuaishuai Tan (Huawei Technologies, China); Chuan Zhao, Lizhi Peng and Bo Yang (University of Jinan, China)

0
As smartphones carry more and more private information, it has become the main target of malware attacks. Threats on mobile devices have become increasingly sophisticated, making it imperative to develop effective tools that are able to detect and counter such threats. Unfortunately, existing malware detection tools based on machine learning techniques struggle to keep up due to the difficulty in performing online incremental update on the detection models. In this paper, a Two-tier Architecture Malware Detection (TAMD) method is proposed, which can learn from the statistical features of network traffic to detect malware. The first layer of TAMD identifies uncertain samples in the training set through a preliminary classification, whereas the second layer builds an improved classifier by filtering out such samples. We enhance TAMD with an incremental leaning based technique (TAMD-IL), which allows to incrementally update the detection models without retraining it from scratch by removing and adding sub-models in TAMD. We experimentally demonstrate that TAMD outperforms the existing methods with up to 98.72% on precision and 96.57% on recall. We also evaluate TAMD-IL on four concept drift datasets and compare it with classical machine learning algorithms, two state-of-the-art malware detection technologies, and three incremental learning technologies. Experimental results show that TAMD-IL is efficient in terms of both update time and memory usage.

Application-Layer DDoS Defense with Reinforcement Learning

Yebo Feng, Jun Li and Thanh Nguyen (University of Oregon, USA)

1
Application-layer distributed denial-of-service (L7 DDoS) attacks, by exploiting application-layer requests to overwhelm functions or components of victim servers, have become a major rising threat to today's Internet. However, because the traffic from an L7 DDoS attack appears legitimate in transport and network layers, it is difficult for traditional DDoS solutions to detect and defend against an L7 DDoS attack.
In this paper, we propose a new, reinforcement-learning-based approach to L7 DDoS attack defense. We introduce a multi-objective reward function to guide a reinforcement learning agent to learn the most suitable action in mitigating L7 DDoS attacks. Consequently, while actively monitoring and analyzing the victim server, the agent can apply different strategies under different conditions to protect the victim: When an L7 DDoS attack is overwhelming, the agent will aggressively mitigate as many malicious requests as possible, thereby keeping the victim server functioning (even at the cost of sacrificing a small number of legitimate requests); otherwise, the agent will conservatively mitigate malicious requests instead, with a focus on minimizing collateral damage to legitimate requests. Our evaluation results show that our approach can mitigate 98.73% of the malicious application messages when the victim is brought to its knees and achieve minimal collateral damage when the L7 DDoS attack is tolerable.

Localizing Failure Root Causes in a Microservice through Causality Inference

Yuan Meng (Tsinghua University, China); Shenglin Zhang and Yongqian Sun (Nankai University, China); Ruru Zhang (NanKai University, China); Zhilong Hu (Nankai University, China); Yiyin Zhang, Chenyang Jia and Zhaogang Wang (Alibaba Group, China); Dan Pei (Tsinghua University, China)

0
An increasing number of Internet applications are applying microservice architecture due to its flexibility and clear logic.
The stability of microservice is thus vitally important for these applications' quality of service.
Accurate failure root cause localization can help operators quickly recover microservice failures and mitigate loss.
Although cross-microservice failure root cause localization has been well studied, how to localize failure root causes in a microservice so as to quickly mitigate this microservice has not yet been studied.
In this work, we propose a framework, MicroCause, to accurately localize the root cause monitoring indicators in a microservice.
MicroCause combines a new pass condition time series (PCTS) algorithm which accurately captures the sequential relationship of time series data, and a novel temporal cause oriented random walk (TCORW) method integrating the causal relationship, temporal order and priority information of monitoring data.
We evaluate MicroCause based on 86 real-world failure cases collected from a top tier global online shopping service.
Our experiments show that the top 5 accuracy AC@5 of MicroCause for intra-microservice failure root cause localization is 98.7%, which is greatly higher (by 33.4%) than the best baseline method.

LogSayer: Log Pattern-driven Cloud Component Anomaly Diagnosis with Machine Learning

Pengpeng Zhou, Yang Wang and Zhenyu Li (Institute of Computing Technology, Chinese Academy of Sciences, China); Xin Wang (Stony Brook University, USA); Gareth Tyson (Queen Mary, University of London, United Kingdom (Great Britain)); Gaogang Xie (Institute of Computing Technology, Chinese Academy of Sciences, China)

0
Anomaly diagnosis is a critical task for building a reliable cloud system and speeding up the system recovery form failures. With the increase of scales and applications of clouds, they are more vulnerable to various anomalies, and it is more challenging for anomaly troubleshooting. System logs that record significant events at critical time points become excellent sources of information to perform anomaly diagnosis. Nevertheless, existing log-based anomaly diagnosis approaches fail to achieve high precision in highly concurrent environments due to interleaved unstructured logs. Besides, transient anomalies that have no obvious features are hard to detect by these approaches. To address this gap, this paper proposes LogSayer, a log pattern-driven anomaly detection model. LogSayer represents the system state by identifying suitable statistical features (such as frequency, surge), which are not sensitive to the exact log sequence. It then measures changes in the log pattern when a transient anomaly occurs. LogSayer uses Long Short-Term Memory (LSTM) neural networks to learn the historical correlation of log patterns and applies a BP neural network for adaptive anomaly decisions. Our experimental evaluations over the HDFS and OpenStack data sets show that LogSayer outperforms the state-of-the-art log-based approaches with precision over 98%.
Session Chair

Yong Cui (Tsinghua U)

Session 3D

Network Optimization and Network Intelligence

Conference
4:30 PM — 6:10 PM HKT
Local
Jun 17 Wed, 1:30 AM — 3:10 AM PDT

Taming the Wildcards: Towards Dependency-free Rule Caching with FreeCache

Rui Li, Bohan Zhao, Ruixin Chen and Jin Zhao (Fudan University, China)

2
Wildcard rules are implemented in various important networking scenarios, including QoS, firewall, access control, and network traffic monitoring and analysis. However, there are cross-rule dependencies between wildcard rules, which both increase significant overhead and affect the semantic correctness of packet classification when caching rules. Considerable efforts have been made to mitigate the impacts of the dependency issue in rule caching, but it is still a bottleneck for cache systems.

In this paper, we show how to give applications the flexibility of completely dependency-free wildcard rule caching by decoupling the cached rules and their dependent rules. Our FreeCache scheme has wide applicability to packet classification devices with wildcard rule caching. We validate the effectiveness of FreeCache through two respects: (1) Implementing various cache algorithms (e.g., LSTM) and cache replacement algorithms (e.g., ARC, LIRS) that are difficult to use in dependency-bound situations in the cache system with FreeCache. (2) Developing a prototype in a Software-Defined Network (SDN), where hybrid OpenFlow switches use TCAM as cache and RAM as auxiliary memory. Our experimental results reveal that FreeCache improves the cache performance by up to 60.88% in the offline scenario. FreeCache also offers the promise of applying any existing caching algorithms to wildcard rule caching while guaranteeing the properties of semantic correctness and equivalence.

Additive and Subtractive Cuckoo Filters

Kun Huang (Southern Universtiy of Science and Technology & Peng Cheng Laboratory, China); Tong Yang (Peking University, China)

2
Bloom filters (BFs) are fast and space-efficient data structures used for set membership queries in many applications. BFs are required to satisfy three key requirements: low space cost, high-speed lookups, and fast updates. Prior works do not satisfy these requirements at the same time. The standard BF does not support deletions of items and the variants that support deletions need additional space or performance overhead. The state-of-the-art cuckoo filters (CF) has high performance with seemingly low space cost. However, the CF suffers a critical issue of varying space cost per item. This is because the exclusive-OR (XOR) operation used by the CF requires the total number of buckets to be a power of two, leading to the space inflation.
To address the issue, in this paper we propose a scalable variant of the cuckoo filter called additive and subtractive cuckoo filter (ASCF). We aim to improve the space efficiency while sustaining comparably high performance. The ASCF uses the addition and subtraction (ADD/SUB) operations instead of the XOR operation to compute an item's two candidate bucket indexes based on its fingerprint. Experimental results show that the ASCF achieves both low space cost and high performance. Compared to the CF, the ASCF reduces up to 1.9x space cost per item while maintaining the same lookup and update throughput. In addition, the ASCF outperforms other filters in both space cost and performance.

Adaptive and Robust Network Routing Based on Deep Reinforcement Learning with Lyapunov Optimization

Zirui Zhuang, Jingyu Wang, Qi Qi and Jianxin Liao (Beijing University of Posts and Telecommunications, China); Zhu Han (University of Houston, USA)

0
The most recent development of the Internet of Things brings massive timely-sensitive and yet bursty data flows. The adaptive network control has been explored using deep reinforcement learning, but it is not sufficient for extremely bursty network traffic flows, especially when the network traffic pattern may change over time. We model the routing control in an environment with time-variant link delays as a Lyapunov optimization problem. We identify that there is a tradeoff between optimization performance and modeling accuracy when the propagation delays are included. We propose a novel deep reinforcement learning-based adaptive network routing method to tackle the issues mentioned above. A Lyapunov optimization technique is used to reduce the upper bound of Lyapunov drift, which leads to improved queuing stability in networked systems. Experiment results show that the proposed method can learn a routing control policy and adapt to the changing environment. The proposed method outperforms the baseline backpressure method in multiple settings, and converges faster than existing methods. Moreover, the deep reinforcement learning module can effectively learn a better estimation of long-term Lyapunov drift and penalty functions, and thus it provides superior results in terms of the backlog size, end-to-end latency, age of information, and throughput. Extensive experiments also show that the proposed model performs well under various topologies, and thus the proposed model can be used in general cases. Also the user can adjust the preference parameter at ant time without the need to retrain the neural networks.

Multi-layer Coordination for High-Performance Energy-Efficient Federated Learning

Li Li (Shenzhen Institutes Of Advanced Technology Chinese Academy Of Sciences, China); Jun Wang (FutureWei Technology, USA); Xu Chen (Sun Yat-sen University, China); Cheng-Zhong Xu (University of Macau, China)

2
Federated Learning is designed for multiple mobile devices to collaboratively train an artificial intelligence model while preserving data privacy. Instead of collecting the raw training data from mobile devices to the cloud, Federated Learning coordinates a group of devices to train a shared model in a distributed manner with the training data located on the devices. However, in order to effectively deploy Federated Learning on resource-constrained mobile devices, several critical issues including convergence rate, scalability and energy efficiency should be well addressed.

In this paper, we propose MCFL, a multi-layer online coordination framework for high-performance energy efficient federated learning. MCFL consists of two layers: a macro-layer on the central server and a micro-layer on each participating device. In each training round, the macro coordinator performs two tasks, namely, selecting the right devices to participate, and estimating
a time limit, such that the overall training time is significantly reduced while still guaranteeing the model accuracy. Unlike existing systems, MCFL removes the restriction that participating devices must be connected to power sources, thus allowing more timely and ubiquitous training. This clearly requires on-device training to be highly energy-efficient. To this end, the micro coordinator determines optimal schedules for hardware resources
in order to meet the time limit set by the macro coordinator with the least amount of energy consumption. Tested on real devices as well as simulation testbed, MCFL has shown to be able to effectively balance the convergence rate, model accuracy and energy efficiency. Compared with existing systems, MCFL can
achieve a speedup up to 8.66 and reduce energy consumption by up to 76.5% during the training process.

Private Deep Neural Network Models Publishing for Machine Learning as a Service

Yunlong Mao, Boyu Zhu, Wenbo Hong, Zhifei Zhu, Yuan Zhang and Sheng Zhong (Nanjing University, China)

2
Machine learning as a service has emerged recently to relieve tensions between heavy deep learning tasks and increasing application demands. A deep learning service provider could help its clients to benefit from deep learning techniques at an affordable price instead of huge resource consumption. However, the service provider may have serious concerns about model privacy when a deep neural network model is published. These concerns are reasonable because recent studies on attacks against deep neural network models (e.g. model inversion attack, membership inference attack) have proved that publishing a private model will cause high risks for model privacy leakage. Previous model publishing solutions mainly depend on additional artificial noise. By adding elaborated noises to parameters or gradients during the training phase, strong privacy guarantees like differential privacy could be achieved. However, this kind of approach cannot give guarantees on some other aspects, such as the quality of the disturbingly trained model and the convergence of the modified learning algorithm. In this paper, we propose an alternative private deep neural network model publishing solution, which caused no interference in the original training phase. We provide privacy, convergence and quality guarantees for the published model at the same time. Furthermore, our solution can achieve a smaller privacy budget when compared with artificial noise based training solutions proposed in previous works. Specifically, our solution gives an acceptable test accuracy with privacy budget epsilon=1. Meanwhile, membership inference attack accuracy will be deceased from nearly 90% to around 60% across all classes.
Session Chair

Fangming Liu (Huazhong UST)

Made with in Toronto · Privacy Policy · © 2022 Duetone Corp.