Y. Thomas, N. Fotiou, Iakovos Pittaras, G. Xylomenos, Spyros Voulgaris, George C. Polyzos
{"title":"Peer Clustering for the InterPlanetary File System","authors":"Y. Thomas, N. Fotiou, Iakovos Pittaras, G. Xylomenos, Spyros Voulgaris, George C. Polyzos","doi":"10.1145/3607504.3609289","DOIUrl":"https://doi.org/10.1145/3607504.3609289","url":null,"abstract":"Distributed Hash Tables (DHT) are once again attracting research interest, 20 years after their inception, as a promising solution for scalable and decentralized content storage. A prominent example is the InterPlanetary File System (IPFS), a distributed peer-to-peer storage system with more than 20K online peers, which uses the Kademlia DHT. We design and implement H-Kademlia, a hierarchical version of Kademlia for IPFS, where peers are distributed into disjoint sets of users, or clusters. Peer clustering can offer resilience to network partition, privacy of in-cluster content lookups, as well as improved caching, content filtering and access control. We assess the performance of IPFS over H-Kademlia via simulations that use real traces from the IPFS network. Our findings show that our design delivers the benefits of peer clustering, without significant performance penalties.","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115242759","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}
{"title":"Performance of Meshed Tree Protocol in Data Center Networks","authors":"P. Willis, Miaoxin Li, N. Shenoy","doi":"10.1145/3607504.3609290","DOIUrl":"https://doi.org/10.1145/3607504.3609290","url":null,"abstract":"Data Center Networks (DCNs) are the communications backbone of data centers. Data centers support thousands of servers, which pose serious DCN architectural and design challenges. Operational demands on DCN protocols to adapt to the new architectures and topologies are met by using multiple off-the-shelf protocols and retrofitting them to the new needs. While major research efforts focus on DCN architectures and topologies, few efforts are directed to address the growing DCN operational and configuration complexity and troubleshooting and management of DCNs. In this work we demonstrate that given the highly structured and symmetrical DCN topologies an extremely simple protocol can replace the multiple protocols that operationally overload DCN routers. We demonstrate our claims by comparing the performance of our proposed Multi Root Meshed Tree Protocol (MR-MTP) to the popular protocol set comprising of Border Gateway Protocol (BGP) for routing, Equal Cost Multipath Protocol (ECMP) for load balancing and Bidirectional Forwarding Detection (BFD) to speed up convergence in a folded-Clos topology. The performance tests were conducted on the Global Environment for Network Innovation (GENI) testbed. Operational MR-MTP C code is available in Github along with all the test scripts. MR-MTP is backward compatible to the Internet Protocol (IP) and Ethernet. From our results it will be clear that we can achieve highly efficient DCN operations and significantly improve performance by rethinking our fundamental approach to routing and IP packet forwarding.","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"33 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124408027","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}
Changgang Zheng, Benjamin Rienecker, Noa Zilberman
{"title":"QCMP: Load Balancing via In-Network Reinforcement Learning","authors":"Changgang Zheng, Benjamin Rienecker, Noa Zilberman","doi":"10.1145/3607504.3609291","DOIUrl":"https://doi.org/10.1145/3607504.3609291","url":null,"abstract":"Traffic load balancing is a long time networking challenge. The dynamism of traffic and the increasing number of different workloads that flow through the network exacerbate the problem. This work presents QCMP, a Reinforcement-Learning based load balancing solution. QCMP is implemented within the data plane, providing dynamic policy adjustment with quick response to changes in traffic. QCMP is implemented using P4 on a switch-ASIC and using BMv2 in a simulation environment. Our results show that QCMP requires negligible resources, runs at line rate, and adapts quickly to changes in traffic patterns.","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126758648","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}
{"title":"Homomorphic Routing: Private Data Forwarding in the Internet","authors":"F. Tusa, David Griffin, M. Rio","doi":"10.1145/3607504.3609287","DOIUrl":"https://doi.org/10.1145/3607504.3609287","url":null,"abstract":"We propose a new private routing and packet forwarding scheme for the Internet---Homomorphic Routing (HR)---that enables endpoints to communicate with one another without divulging source or destination addresses to the routers or service providers along the path. This is achieved via homomorphic encryption, whereby domains can match encrypted address ranges with encrypted destinations of packets without the need of decryption. Compared to approaches such as source or onion routing, HR is a hop-by-hop solution that allows current BGP-like decisions and traffic engineering techniques to remain largely unchanged, while per-flow state need not be maintained by routers. Preliminary performance evaluation shows that HR implies a tolerable computational overhead compared to plain text operations. Through aggregation we can compress inter-domain routing rules to around 5% of those required for current IPv6 and we can organize encrypted forwarding rules so that matching can be achieved in logarithmic time.","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132404007","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}
Mircea M. Iordache-Sica, Tula Kraiser, O. Komolafe
{"title":"Seamless Hardware-Accelerated Kubernetes Networking","authors":"Mircea M. Iordache-Sica, Tula Kraiser, O. Komolafe","doi":"10.1145/3607504.3609292","DOIUrl":"https://doi.org/10.1145/3607504.3609292","url":null,"abstract":"Kubernetes has quickly become the industry-standard Service management and orchestration framework. However, Kubernetes networking has traditionally been achieved by software implementations of CNI (Container Networking Interface) plugins with performance limitations. This paper puts forward the integration of hardware networking appliances for enhancing Kubernetes performance in Data Center environments by utilising the Hardware Accelerated Kubernetes (HAcK) architecture. The proposed architecture is applied to the implementation of the HAcK LoadBalancer which leverages industry-grade network appliances for creating unified application and infrastructure management mechanisms. Performance of the system is evaluated in real data centre scenarios.","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131476796","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}
{"title":"If Iterative Diffusion Is The Answer, What Was The Question?","authors":"David Guzman, D. Trossen, J. Ott","doi":"10.1145/3607504.3609288","DOIUrl":"https://doi.org/10.1145/3607504.3609288","url":null,"abstract":"A Distributed Concensus System (DCS) is a key component for a distributed system, maintaining and consenting over distributed state, as often required in distributed filesystems, cryptocurrencies and other applications. The prevalent mechanism to distribute the information needed to reach consensus in a DCS is through an iterative diffusion with the goal of distributing the (updated) information to at least the majority of DCS participants. Such diffusion is not just costly in maintaining the needed communication relations but also incurs latency in performing the diffusion itself. When looking closer at the goal of achieving the majority rule, however, one may wonder of the lack of multicast being used for achieving the same goal. This paper thus returns to the original intention of how to distribute information to a large set of users and outlines reasons for why multicast has not been widely used thus far. We evaluate the potential gain in doing so, based on an analytical model for the diffusion latency in an iterative diffusion DCS, bounded through empirical studies in an Ethereum system.","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128766586","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}
{"title":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","authors":"","doi":"10.1145/3607504","DOIUrl":"https://doi.org/10.1145/3607504","url":null,"abstract":"","PeriodicalId":308036,"journal":{"name":"Proceedings of the 2nd ACM SIGCOMM Workshop on Future of Internet Routing & Addressing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124338247","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}