Measuring Content and Performance of IPFS Network

doi:10.3778/j.issn.1002-8331.2102-0122

Abstract

Abstract: Interplanetary file system（IPFS） is a decentralized file system. Since it can serve as a storage solution for various blockchain systems, it has been used in many applications. However, there remains questions about IPFS itself, including what kinds of contents have been stored on the system, and what are the key factors that affect performance of IPFS. By implementing and deploying a crawler in the IPFS network, measurement is done about both the content and the performance aspects of IPFS. Content measurement results show that more than 80% of files stored on IPFS are less than 1?MB. The graph comprising files and directories has a degree distribution that approximately follows a power law. In performance measurement, the IPFS network is found to have around 15,000 peers, which forms a connectivity graph whose in-degree follows a power law distribution. Chunk size and number of chunks have a direct impact on performance. Considerable difference in DHT resolution time is found between private and public network modes. DHT resolution is expected to be faster in both modes if the file being resolved has more providers.

Key words: interplanetary file system（IPFS）, peer-to-peer network, network measurement, power law distribution

摘要： IPFS（interplanetary file system，星际文件系统）是一个去中心化的分布式文件系统，作为对区块链系统在存储方面的重要拓展获得许多应用。然而，对于IPFS本身尚有许多问题亟待回答，包括IPFS被用于存储怎样的内容、影响性能的关键因素等。通过在IPFS网络中部署爬虫程序，收集数据，从内容和性能两方面进行了测量研究。内容测量的结果发现，IPFS网络中存储的文件有80%以上是小于1?MB的小文件；IPFS中存储的文件和目录的构成的图的度近似满足幂律分布。在性能测量中，发现当前IPFS网络由约15?000个节点组成，节点互相连接构成的网络的入度满足幂律分布。在IPFS中存储文件时，分片大小和数量对性能有直接的影响，而在不同的网络环境中，即私有集群和公开网络中，DHT查询的性能相差较大；提供者数量越多，DHT解析越快。

关键词: 星际文件系统（IPFS）, 点对点网络, 网络测量, 幂律分布

DING Bowen, XU Yuedong, WANG Liang. Measuring Content and Performance of IPFS Network[J]. Computer Engineering and Applications, 2022, 58(7): 97-105.

丁博文, 徐跃东, 王亮. IPFS网络内容和性能测量[J]. 计算机工程与应用, 2022, 58(7): 97-105.

References

[1] ZHENG Z，XIE S，DAI H，et al.An overview of blockchain technology：Architecture，consensus，and future trends[C]//Proceedings of 2017 IEEE International Congress on Big Data（BigData Congress），Honolulu，USA，2017.
[2] BENET J.IPFS-content addressed，versioned，P2P file system[J].arXiv：1407.3561，2014.
[3] ALI M S，DOLUI K，AN-TONELLI F.IoT data privacy via blockchains and IPFS[C]//Proceedings of the 7th International Conference on the Internet of Things（IoT’17），New York，USA，2017.
[4] CONFAIS B，LEBRE A，PARREIN B.An object store service for a fog/edge computing infrastructure based on IPFS and a scale-out NAS[C]//Proceedings of 2017 IEEE 1st International Conference on Fog and Edge Computing（ICFEC），2017.
[5] SUN J，YAO X，WANG S，et al.Blockchain-based secure storage and access scheme for electronic medical records in IPFS[J].IEEE Access，2020，8：59389-59401.
[6] NIZAMUDDIN N，SALAH K，AJMAL AZAD K，et al.Decentralized document version control using ethereum blockchain and IPFS[J].Computers & Electrical Engineering，2019，76（6）：183-197.
[7] BAUMGART I，MIES S.S.Kademlia：A practicable approach towards secure key-based routing[C]//Proceedings of 2007 International Conference on Parallel and Distributed Systems，2007.
[8] MERKLE R.A digital signature based on a conventional encryption function[C]//Advances in Cryptology—CRYPTO’87，1987.
[9] MAYMOUNKOV P，MAZIèRES D.Kademlia：A peer-to-peer information system based on the XOR metric[C]//Proceedings of IPTPS 2002：Peer-to-Peer Systems，2002.
[10] BIGCHAINDB GMBH.BigchainDB 2.0 the blockchain database[R/OL]（2018?05）[2021?02?04].https：//www.bigchaindb.com/whitepaper/bigchaindb-whitepaper.pdf.
[11] ALAM S，KELLY M，NELSON M L.Interplanetary Wayback：The permanent web archive[C]//Proceedings of 2016 IEEE/ACM Joint Conference on Digital Libraries（JCDL），Newark，USA，2016.
[12] POUWELSE J，GARBACKI P，EPEMA D，et al.The bittorrent P2P file-sharing system：Measurements and analysis[C]//Proceedings of IPTPS 2005：Peer-to-Peer Systems IV，2005.
[13] GUO L，CHEN S，XIAO Z，et al.Measurements，analysis，and modeling of BitTorrent-like systems[C]//Proceedings of the 5th ACM SIGCOMM Conference on Internet Measurement，Berkeley，USA，2005.
[14] YU J，LI Z.Active measurement of routing table in kad[C]//Proceedings of 2009 6th IEEE Consumer Communications and Networking Conference，Las Vegas，USA，2009.
[15] SHEN J，LI Y，ZHOU Y，et al.Understanding I/O performance of IPFS storage：A client’s perspective[C]//Proceedings of the International Symposium on Quality of Service，New York，USA，2019.
[16] HENNINGSEN S，FLORIAN M，RUST S，et al.Mapping the interplanetary filesystem[C]//Proceedings of 2020 IFIP Networking Conference（Networking），Paris，France，2020.
[17] BARABáSI A L.Scale-free networks：A decade and beyond[J].Science，2009，25：412-413.
[18] CLAUSET A，SHALIZI C R，NEWMAN M E J.Power-law distributions in empirical data[J].SIAM Review，2009，51（4）：661-703.