Distributed machine learning is a promising machine learning approach for academia and industry. It can generate a machine learning model for dispersed training data via iterative training in a distributed fashion. To speed up the training process of distributed machine learning, it is essential to reduce the communication load among training nodes. In this paper, we propose a layer-wise gradient compression scheme based on principal component analysis and error accumulation. The key of our solution is to consider the gradient characteristics and architecture of neural networks by taking advantage of the compression ability enabled by PCA and the feedback ability enabled by error accumulation. The preliminary results on image classification task show that our scheme achieves good performance and reduces 97% of the gradient transmission.
{"title":"Efficient and Structural Gradient Compression with Principal Component Analysis for Distributed Training","authors":"Jiaxin Tan, Chao Yao, Zehua Guo","doi":"10.1145/3600061.3603140","DOIUrl":"https://doi.org/10.1145/3600061.3603140","url":null,"abstract":"Distributed machine learning is a promising machine learning approach for academia and industry. It can generate a machine learning model for dispersed training data via iterative training in a distributed fashion. To speed up the training process of distributed machine learning, it is essential to reduce the communication load among training nodes. In this paper, we propose a layer-wise gradient compression scheme based on principal component analysis and error accumulation. The key of our solution is to consider the gradient characteristics and architecture of neural networks by taking advantage of the compression ability enabled by PCA and the feedback ability enabled by error accumulation. The preliminary results on image classification task show that our scheme achieves good performance and reduces 97% of the gradient transmission.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133401438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In current high-speed data center networks, congestion control is crucial for ensuring consistent high performance. Over the past decade, researchers and developers have explored several congestion signals such as ECN, RTT, and INT. However, most of the existing congestion control algorithms suffer from either imprecise congestion detection due to ambiguous signals or excessive bandwidth loss due to aggressive rate decrease. This paper proposes a novel congestion control mechanism called TOD, which is a trend-oriented delay-based approach designed for lossless data center networks. TOD leverages the change in RTT to learn the congestion trend and adjusts the sending rate accordingly. By analyzing the congestion trend, the sender reacts by adjusting the sending rate to a reasonable level, while still maintaining high bandwidth utilization to dismiss congestion. The sender uses a reference rate, which is calculated by the receiver and communicated back to the sender, to achieve this target. Therefore, TOD is a sender-receiver cooperative congestion control mechanism. We evaluate TOD extensively in NS-3 simulations using both microbenchmark and macrobenchmark. Our experiments demonstrate that TOD outperforms DCQCN and Timely in terms of FCT and convergence speed.
{"title":"TOD: Trend-Oriented Delay-Based Congestion Control in Lossless Datacenter Network","authors":"Kaixin Huang, Wentao Liu, Weihang Li, Lang Cheng","doi":"10.1145/3600061.3600079","DOIUrl":"https://doi.org/10.1145/3600061.3600079","url":null,"abstract":"In current high-speed data center networks, congestion control is crucial for ensuring consistent high performance. Over the past decade, researchers and developers have explored several congestion signals such as ECN, RTT, and INT. However, most of the existing congestion control algorithms suffer from either imprecise congestion detection due to ambiguous signals or excessive bandwidth loss due to aggressive rate decrease. This paper proposes a novel congestion control mechanism called TOD, which is a trend-oriented delay-based approach designed for lossless data center networks. TOD leverages the change in RTT to learn the congestion trend and adjusts the sending rate accordingly. By analyzing the congestion trend, the sender reacts by adjusting the sending rate to a reasonable level, while still maintaining high bandwidth utilization to dismiss congestion. The sender uses a reference rate, which is calculated by the receiver and communicated back to the sender, to achieve this target. Therefore, TOD is a sender-receiver cooperative congestion control mechanism. We evaluate TOD extensively in NS-3 simulations using both microbenchmark and macrobenchmark. Our experiments demonstrate that TOD outperforms DCQCN and Timely in terms of FCT and convergence speed.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"229 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130896733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Multicast is widely used to support a huge range of applications with one-to-many or many-to-many communication patterns. However, multicast systems do not scale due to considerable state and communication overheads. Some stateful multicast approaches require maintaining the state of each multicast session at switches, thus incurring large memory overhead. Some stateless ones utilize Bloom filter (BF) to encode multicast tree into the packet header to minimize communication overhead, but potentially suffer from the substantial false positive due to the probabilistic nature of Bloom filter. In this paper, we propose a stateless multicast scheme MEB, which uses Bloom filter to achieve large-scale multicast communication with low error, small overhead and high scalability. Specifically, to control the rate of false positive, MEB elaborately selects the hash functions for Bloom filters when constructing the packet header at the sender side, and makes forwarding decision according to packet header at the switch with negligible overhead. We compare MEB against the state-of-the-art multicast system in large-scale simulations. The test results show that MEB reduces the traffic overhead by up to 70% with small error rate.
{"title":"MEB: an Efficient and Accurate Multicast using Bloom Filter with Customized Hash Function","authors":"Zihao Chen, Jiawei Huang, Qile Wang, Jingling Liu, Zhaoyi Li, Shengwen Zhou, Zhidong He","doi":"10.1145/3600061.3600062","DOIUrl":"https://doi.org/10.1145/3600061.3600062","url":null,"abstract":"Multicast is widely used to support a huge range of applications with one-to-many or many-to-many communication patterns. However, multicast systems do not scale due to considerable state and communication overheads. Some stateful multicast approaches require maintaining the state of each multicast session at switches, thus incurring large memory overhead. Some stateless ones utilize Bloom filter (BF) to encode multicast tree into the packet header to minimize communication overhead, but potentially suffer from the substantial false positive due to the probabilistic nature of Bloom filter. In this paper, we propose a stateless multicast scheme MEB, which uses Bloom filter to achieve large-scale multicast communication with low error, small overhead and high scalability. Specifically, to control the rate of false positive, MEB elaborately selects the hash functions for Bloom filters when constructing the packet header at the sender side, and makes forwarding decision according to packet header at the switch with negligible overhead. We compare MEB against the state-of-the-art multicast system in large-scale simulations. The test results show that MEB reduces the traffic overhead by up to 70% with small error rate.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122140104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tian Pan, Shuihai Hu, Guangyu An, Xincai Fei, Fanzhao Wang, Yueke Chi, Minglan Gao, Hao Wu, Jiao Zhang, Tao Huang, Jingbin Zhou, Kun Tan
TCP was designed to provide stream-oriented communication service for bulk data transfer applications (e.g., FTP and Email). With four-decade development, Internet applications have undergone significant changes, which now involve highly dynamic traffic pattern and message-oriented communication paradigm. However, the impact of this substantial evolution on congestion control (CC) has not been fully studied. Most of the network transports today still make the long-held assumption about application traffic, i.e., a byte stream with an unlimited data arrival rate. In this paper, we demonstrate, through both analyses and experiments, that the emerging traffic dynamics and message-level data structure have huge impacts on the correctness and effectiveness of CC, but none of the existing solutions treats these two characteristics appropriately. Therefore, we present Amphis, a new CC framework that re-architects the current pure network-oriented design into a dual-control architecture, which combines application-coordinated control and network-oriented control. Amphis contains two novel ideas, i.e., pattern-driven proactive probing for handling traffic dynamics and message-driven adaptive optimization for optimizing message transmission performance. Our preliminary results show that Amphis holds great promise in terms of accurate bandwidth estimation under dynamic traffic conditions and effective data transfer at message granularity.
{"title":"Amphis: Rearchitecturing Congestion Control for Capturing Internet Application Variety","authors":"Tian Pan, Shuihai Hu, Guangyu An, Xincai Fei, Fanzhao Wang, Yueke Chi, Minglan Gao, Hao Wu, Jiao Zhang, Tao Huang, Jingbin Zhou, Kun Tan","doi":"10.1145/3600061.3600076","DOIUrl":"https://doi.org/10.1145/3600061.3600076","url":null,"abstract":"TCP was designed to provide stream-oriented communication service for bulk data transfer applications (e.g., FTP and Email). With four-decade development, Internet applications have undergone significant changes, which now involve highly dynamic traffic pattern and message-oriented communication paradigm. However, the impact of this substantial evolution on congestion control (CC) has not been fully studied. Most of the network transports today still make the long-held assumption about application traffic, i.e., a byte stream with an unlimited data arrival rate. In this paper, we demonstrate, through both analyses and experiments, that the emerging traffic dynamics and message-level data structure have huge impacts on the correctness and effectiveness of CC, but none of the existing solutions treats these two characteristics appropriately. Therefore, we present Amphis, a new CC framework that re-architects the current pure network-oriented design into a dual-control architecture, which combines application-coordinated control and network-oriented control. Amphis contains two novel ideas, i.e., pattern-driven proactive probing for handling traffic dynamics and message-driven adaptive optimization for optimizing message transmission performance. Our preliminary results show that Amphis holds great promise in terms of accurate bandwidth estimation under dynamic traffic conditions and effective data transfer at message granularity.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127933221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hongyun Wu, Qi Ling, Penghui Mi, Chaoyang Ji, Yinliang Hu, Yibo Pi
The massiveness of the Internet makes it rather difficult to achieve high-coverage monitoring at scale with reasonable overhead. The traditional wisdom for scalable and high-coverage Internet monitoring is to consider clients in each /24 as a whole and only monitor the representatives, either by active probing or by passive traffic sniffing, such that performance of the rest can be predicted for high coverage. There are two basic assumptions behind this traditional wisdom: 1) clients in the same /24 have similar performance, and 2) tracking all targeted /24s equates to full-coverage monitoring. With the increasing prevalence of load balancing, both assumptions are now questionable. Through large-scale measurements, we evaluate the coverage and predictability issues of current practices, motivate the necessity of link-level fine-grained, high-coverage monitoring, and present new insights on how to achieve it. Our key findings are: 1) the current practices using the representatives of /24s may fail to capture the changes of up to 85% of links in the Internet; 2) the path difference between client flows to the same /24 is both significant and prevalent; 3) it is possible to cover most of the visible links from DCs to both small and large prefixes by carefully choosing client flows; 4) high-coverage monitoring can be achieved with at least three times less overhead than direct link monitoring.
{"title":"Towards Fine-Grained, High-Coverage Internet Monitoring at Scale","authors":"Hongyun Wu, Qi Ling, Penghui Mi, Chaoyang Ji, Yinliang Hu, Yibo Pi","doi":"10.1145/3600061.3600085","DOIUrl":"https://doi.org/10.1145/3600061.3600085","url":null,"abstract":"The massiveness of the Internet makes it rather difficult to achieve high-coverage monitoring at scale with reasonable overhead. The traditional wisdom for scalable and high-coverage Internet monitoring is to consider clients in each /24 as a whole and only monitor the representatives, either by active probing or by passive traffic sniffing, such that performance of the rest can be predicted for high coverage. There are two basic assumptions behind this traditional wisdom: 1) clients in the same /24 have similar performance, and 2) tracking all targeted /24s equates to full-coverage monitoring. With the increasing prevalence of load balancing, both assumptions are now questionable. Through large-scale measurements, we evaluate the coverage and predictability issues of current practices, motivate the necessity of link-level fine-grained, high-coverage monitoring, and present new insights on how to achieve it. Our key findings are: 1) the current practices using the representatives of /24s may fail to capture the changes of up to 85% of links in the Internet; 2) the path difference between client flows to the same /24 is both significant and prevalent; 3) it is possible to cover most of the visible links from DCs to both small and large prefixes by carefully choosing client flows; 4) high-coverage monitoring can be achieved with at least three times less overhead than direct link monitoring.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130316555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Trigger-Action Program (TAP) is a popular and significant form of Internet of Things (IoT) applications, commonly utilized in smart homes. Existing works either just perform actions based on commands or require human intervention to generate TAPs. With the emergence of Large Language Models (LLMs), it becomes possible for users to create IoT TAPs in zero-code manner using natural language. Thus, we propose ChatIoT, which employs LLMs to process natural language in chats and realizes the zero-code generation of TAPs for existing devices.
{"title":"ChatIoT: Zero-code Generation of Trigger-action Based IoT Programs with ChatGPT","authors":"Fu Li, Jiaming Huang, Yi Gao, Wei Dong","doi":"10.1145/3600061.3603141","DOIUrl":"https://doi.org/10.1145/3600061.3603141","url":null,"abstract":"Trigger-Action Program (TAP) is a popular and significant form of Internet of Things (IoT) applications, commonly utilized in smart homes. Existing works either just perform actions based on commands or require human intervention to generate TAPs. With the emergence of Large Language Models (LLMs), it becomes possible for users to create IoT TAPs in zero-code manner using natural language. Thus, we propose ChatIoT, which employs LLMs to process natural language in chats and realizes the zero-code generation of TAPs for existing devices.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129817148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhen Tian, Lingen Ding, Qing Wang, Desheng Sun, Yunlong Li
Recent years have seen a significant increase in demand for security in Ethernet. In this paper, we propose a novel physical layer security architecture (PHYSec) for Ethernet, operating at the low layer of physical coding sublayer (PCS). Compared to existing encryption schemes, PHYSec provides the following benefits: 1) low overhead due to constructing encryption objects without the restriction of frame size and utilizing native markers to carry security parameters; 2) high security for completely hiding the traffic pattern.
{"title":"PHYSec: A Novel Physical Layer Security Architecture for Ethernet","authors":"Zhen Tian, Lingen Ding, Qing Wang, Desheng Sun, Yunlong Li","doi":"10.1145/3600061.3603133","DOIUrl":"https://doi.org/10.1145/3600061.3603133","url":null,"abstract":"Recent years have seen a significant increase in demand for security in Ethernet. In this paper, we propose a novel physical layer security architecture (PHYSec) for Ethernet, operating at the low layer of physical coding sublayer (PCS). Compared to existing encryption schemes, PHYSec provides the following benefits: 1) low overhead due to constructing encryption objects without the restriction of frame size and utilizing native markers to carry security parameters; 2) high security for completely hiding the traffic pattern.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125607164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bojie Li, Zihao Xiang, Xiaoliang Wang, Hang Ruan, Jingbin Zhou, Kun Tan
Polling and interrupt has long been a trade-off in RDMA systems. Polling has lower latency but each CPU core can only run one thread. Interrupt enables time sharing among multiple threads but has higher latency. Many applications such as databases have hundreds of threads, which is much larger than the number of cores. So, they have to use interrupt mode to share cores among threads, and the resulting RDMA latency is much higher than the hardware limits. In this paper, we analyze the root cause of high costs in RDMA interrupt delivery, and present FastWake, a practical redesign of interrupt-mode RDMA host network stack using commodity RDMA hardware, Linux OS, and unmodified applications. Our first approach to fast thread wake-up completely removes interrupts. We design a per-core dispatcher thread to poll all the completion queues of the application threads on the same core, and utilize a kernel fast path to context switch to the thread with an incoming completion event. The approach above would keep CPUs running at 100% utilization, so we design an interrupt-based approach for scenarios with power constraints. Observing that waking up a thread on the same core as the interrupt is much faster than threads on other cores, we dynamically adjust RDMA event queue mappings to improve interrupt core affinity. In addition, we revisit the kernel path of thread wake-up, and remove the overheads in virtual file system (VFS), locking, and process scheduling. Experiments show that FastWake can reduce RDMA latency by 80% on x86 and 77% on ARM at the cost of < 30% higher power utilization than traditional interrupts, and the latency is only 0.3 ∼ 0.4 μ s higher than the limits of underlying hardware. When power saving is desired, our interrupt-based approach can still reduce interrupt-mode RDMA latency by 59% on x86 and 52% on ARM.
{"title":"FastWake: Revisiting Host Network Stack for Interrupt-mode RDMA","authors":"Bojie Li, Zihao Xiang, Xiaoliang Wang, Hang Ruan, Jingbin Zhou, Kun Tan","doi":"10.1145/3600061.3600063","DOIUrl":"https://doi.org/10.1145/3600061.3600063","url":null,"abstract":"Polling and interrupt has long been a trade-off in RDMA systems. Polling has lower latency but each CPU core can only run one thread. Interrupt enables time sharing among multiple threads but has higher latency. Many applications such as databases have hundreds of threads, which is much larger than the number of cores. So, they have to use interrupt mode to share cores among threads, and the resulting RDMA latency is much higher than the hardware limits. In this paper, we analyze the root cause of high costs in RDMA interrupt delivery, and present FastWake, a practical redesign of interrupt-mode RDMA host network stack using commodity RDMA hardware, Linux OS, and unmodified applications. Our first approach to fast thread wake-up completely removes interrupts. We design a per-core dispatcher thread to poll all the completion queues of the application threads on the same core, and utilize a kernel fast path to context switch to the thread with an incoming completion event. The approach above would keep CPUs running at 100% utilization, so we design an interrupt-based approach for scenarios with power constraints. Observing that waking up a thread on the same core as the interrupt is much faster than threads on other cores, we dynamically adjust RDMA event queue mappings to improve interrupt core affinity. In addition, we revisit the kernel path of thread wake-up, and remove the overheads in virtual file system (VFS), locking, and process scheduling. Experiments show that FastWake can reduce RDMA latency by 80% on x86 and 77% on ARM at the cost of < 30% higher power utilization than traditional interrupts, and the latency is only 0.3 ∼ 0.4 μ s higher than the limits of underlying hardware. When power saving is desired, our interrupt-based approach can still reduce interrupt-mode RDMA latency by 59% on x86 and 52% on ARM.","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124308216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We study how to reduce the reconfiguration time in hybrid optical-electrical Datacenter Networks (DCNs). With a layer of Optical Circuit Switches (OCSes), hybrid optical-electrical DCNs could reconfigure their logical topologies to better match the on-going traffic patterns, but the reconfiguration time could directly affect the benefits of reconfigurability. The reconfiguration time consists of the topology solver running time and the network convergence time after triggering reconfiguration. However, existing topology solvers either incur high algorithmic complexity or fail to minimize the reconfiguration overhead. In this paper, we propose a novel algorithm that combines the ideas of bipartition and Minimum Cost Flow (MCF) to reduce the overall reconfiguration time. For the first time, we formulate the topology solving problem as an MCF problem with piecewise cost, which strikes a better balance between solver complexity and solution optimality. Our evaluation shows that our algorithm can significantly reduce the network convergence time while consuming less topology solver running time, making its overall performance superior to existing algorithms. Our code and test cases are available at a public repository [25].
{"title":"Reducing Reconfiguration Time in Hybrid Optical-Electrical Datacenter Networks","authors":"Shuyuan Zhang, Shu Shan, Shizhen Zhao","doi":"10.1145/3600061.3600071","DOIUrl":"https://doi.org/10.1145/3600061.3600071","url":null,"abstract":"We study how to reduce the reconfiguration time in hybrid optical-electrical Datacenter Networks (DCNs). With a layer of Optical Circuit Switches (OCSes), hybrid optical-electrical DCNs could reconfigure their logical topologies to better match the on-going traffic patterns, but the reconfiguration time could directly affect the benefits of reconfigurability. The reconfiguration time consists of the topology solver running time and the network convergence time after triggering reconfiguration. However, existing topology solvers either incur high algorithmic complexity or fail to minimize the reconfiguration overhead. In this paper, we propose a novel algorithm that combines the ideas of bipartition and Minimum Cost Flow (MCF) to reduce the overall reconfiguration time. For the first time, we formulate the topology solving problem as an MCF problem with piecewise cost, which strikes a better balance between solver complexity and solution optimality. Our evaluation shows that our algorithm can significantly reduce the network convergence time while consuming less topology solver running time, making its overall performance superior to existing algorithms. Our code and test cases are available at a public repository [25].","PeriodicalId":228934,"journal":{"name":"Proceedings of the 7th Asia-Pacific Workshop on Networking","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129972873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：