{"title":"再现“重构端点拥塞控制”的关键结果","authors":"Tushar Dhoot, Joe Zhang","doi":"10.1109/cps-iotbench56135.2022.00010","DOIUrl":null,"url":null,"abstract":"This work is a reproduction of Narayan et. al’s paper [8] which introduced the Congestion Control Plane (CCP) as a mechanism to develop congestion control algorithms. CCP can be leveraged to improve TCP/IP for Internet of Things (IoT) networks and is therefore relevant in the conversation around the suitability of TCP/IP for IoT applications as described by Dunkels et. al [4].Our reproduction experiments confirm CCP has high fidelity with congestion control implemented directly in the OS kernel. With one exception, our study also confirms that CCP algorithms do not materially differ in their performance. These results affirm that CCP is a promising enhancement over kernel level congestion control.Our experiments did find areas of concern. We discovered several inoperable artifacts in the code provided by the authors. Our experimental setup surfaced issues with reproducing the original results due to changes in the underlying systems. We discovered a graph that was accidentally mislabeled in a material way. Finally, we found a small subset of performance measurements could not be reproduced and flagged these for further research. We also proposed improvements to how replication environments are shared as a general suggestion to improve reproducibility.","PeriodicalId":371398,"journal":{"name":"2022 Workshop on Benchmarking Cyber-Physical Systems and Internet of Things (CPS-IoTBench)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Reproducing Key Results from \\\"Restructuring Endpoint Congestion Control\\\"\",\"authors\":\"Tushar Dhoot, Joe Zhang\",\"doi\":\"10.1109/cps-iotbench56135.2022.00010\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This work is a reproduction of Narayan et. al’s paper [8] which introduced the Congestion Control Plane (CCP) as a mechanism to develop congestion control algorithms. CCP can be leveraged to improve TCP/IP for Internet of Things (IoT) networks and is therefore relevant in the conversation around the suitability of TCP/IP for IoT applications as described by Dunkels et. al [4].Our reproduction experiments confirm CCP has high fidelity with congestion control implemented directly in the OS kernel. With one exception, our study also confirms that CCP algorithms do not materially differ in their performance. These results affirm that CCP is a promising enhancement over kernel level congestion control.Our experiments did find areas of concern. We discovered several inoperable artifacts in the code provided by the authors. Our experimental setup surfaced issues with reproducing the original results due to changes in the underlying systems. We discovered a graph that was accidentally mislabeled in a material way. Finally, we found a small subset of performance measurements could not be reproduced and flagged these for further research. We also proposed improvements to how replication environments are shared as a general suggestion to improve reproducibility.\",\"PeriodicalId\":371398,\"journal\":{\"name\":\"2022 Workshop on Benchmarking Cyber-Physical Systems and Internet of Things (CPS-IoTBench)\",\"volume\":\"56 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-05-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 Workshop on Benchmarking Cyber-Physical Systems and Internet of Things (CPS-IoTBench)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/cps-iotbench56135.2022.00010\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 Workshop on Benchmarking Cyber-Physical Systems and Internet of Things (CPS-IoTBench)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/cps-iotbench56135.2022.00010","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Reproducing Key Results from "Restructuring Endpoint Congestion Control"
This work is a reproduction of Narayan et. al’s paper [8] which introduced the Congestion Control Plane (CCP) as a mechanism to develop congestion control algorithms. CCP can be leveraged to improve TCP/IP for Internet of Things (IoT) networks and is therefore relevant in the conversation around the suitability of TCP/IP for IoT applications as described by Dunkels et. al [4].Our reproduction experiments confirm CCP has high fidelity with congestion control implemented directly in the OS kernel. With one exception, our study also confirms that CCP algorithms do not materially differ in their performance. These results affirm that CCP is a promising enhancement over kernel level congestion control.Our experiments did find areas of concern. We discovered several inoperable artifacts in the code provided by the authors. Our experimental setup surfaced issues with reproducing the original results due to changes in the underlying systems. We discovered a graph that was accidentally mislabeled in a material way. Finally, we found a small subset of performance measurements could not be reproduced and flagged these for further research. We also proposed improvements to how replication environments are shared as a general suggestion to improve reproducibility.