Decentralized Cloud Comparison: Storj & Archon Cloud
In today’s data-driven society, our reliance on cloud storage is now greater than ever. It’s more amazing that the average user of these services don’t realize how cloud works.
In 2012, a survey of more than 1,000 adults was conducted by Wakefield Research showing that nearly one third say they see the cloud as a thing of the future, yet 97% are actually using it and one in five of those Americans (22%) admit that they’ve pretended to know what the cloud is or how it works.
There are a variety of reasons to consider decentralized storage, including speed, reliability, improved security, and reduced costs. This article will highlight and compare two promising projects in the decentralized cloud storage space — Archon and Storj.
Both projects aim to develop the technology enabling decentralized cloud storage — however, there are some key differences in their miner selection process, payment process, file encoding, and general business strategy.
Comparison Table:
Miner Selection Process
Both Storj and Archon have methods to ensure files are stored on high-quality nodes. The key difference is that Storj utilizes randomization with a reputation system whereas Archon utilizes a marketplace with staking mechanisms.
All Storj storage nodes must go through a vetting process becoming fully active. Initially unvetted, all storage nodes are slowly incorporated to store small amounts of data. This is done by “Satellites”. Once a node has proven its ability to successfully store data, it becomes included into the regular potential pool where they are pseudo-randomly selected by the network to store data. A downside to this way of vetting is its vulnerability to “the Honest Geppetto attack ,” where an attacker pretends to be honest until vetted, but once vetted, they hold the data hostage or leave the network while still being paid.
All Archon storage nodes must also go through a vetting process. Storage Providers must first pass a “Proof of Space” before becoming an active participant in the marketplace. In the marketplace, an uploader can filter (based on price, location, and availability of edge nodes) all available hosts and choose the best ones that match their desired level of service. To ensure the security and reliability of the Archon network, all storage providers must also “stake tokens” correlated to the amount of data they wish to store. If a miner fails to provide the level of service they promised in a Service-Level Agreement (SLA), they will lose a portion of their stake. This builds better incentivizes for “good behavior” into the system, and prevent users from abusing the system.
Payment Process
All payments in Storj must go through Satellites (currently centralized) whereas Archon allows direct payment between uploaders and storage miners.
In Storj, all payments made by clients are directed to Satellites, and uploaders may pay with the STORJ Token, credit card, or invoice. The Satellite, in turn, pays storage nodes in STORJ tokens. While most networks handled payments in hard-coded contracts, Storj defaults in assuming that the data will last indefinitely. At the end of each payment period, a Satellite will calculate earnings for each of its storage nodes. Satellites pay storage nodes for the data they store and for piece downloads (month-by-month). As all payments must go through Satellites, they highly recommend you utilize a Satellite that is hosted by either Storj or a trustworthy 3rd party.
In Archon, all payments made by uploaders are paid directly to storage miners with Archon tokens, or through an Archon Integration Node (AIN). AINs make it easy for an uploader from almost any external chain to use the Archon Cloud without having to directly interact with the Archon Cloud Blockchain. As long as the external chain supports smart contracts, an uploader can pay the AIN in their external blockchain’s native token. The AIN will, in turn, pay the appropriate amount of Archon tokens to the storage miners. So far, Archon has built out an Integration with NEO, and has plans to integrate other public chains in the near future. In addition, Archon has plans in the future to allow payments through more traditional means such as credit cards or PayPal.
File Encoding Process
Both Archon and Storj utilize erasure encoding and file sharding to help distribute the files to their miners/storage providers. However, there are some key differences between these two companies’ metadata management and encoding processes.
Primer on Erasure Encoding
The general erasure encoding process is taking a file, encoding it with erasure codes, splitting it into thousands of pieces, and distributing them to various unrelated nodes. Erasure encoding enables the ability to recreate an original file with any subset of these distributed pieces (data shards) as long as one has a specific threshold of data shards. These systems theoretically make it extremely difficult to cause any meaningful disruptions — attempting to maliciously attack the network could cost a would-be attacker upwards of millions of dollars.
Displayed below is a side by side comparison of how these two companies encode their files.
Storj stores a file’s metadata with their Satellites. Storj’s reliance on Satellites adds additional risk to the file retrieval process, as if the Satellite goes down, the file can no longer be retrieved. Storj uses the same type of standard erasure encoding for all files in its system that are used by many projects.
Archon stores file metadata directly in the Archon Datashards. Archon does not use the common encoding library, but rather has developed proprietary erasure encoding algorithms optimized for improved file reliability and performance. Archon also allows for a dynamic choice between different erasure codes, different file sizes and varying retrievability requirements would use the most appropriate encoding algorithm.
Business Strategy & State of Development
Storj is much farther along in terms of funding, development, and community development. Raising $30 MM in an ICO in 2017, they have done a lot of work in the enterprise space. With their Open Source Partnership Program, they have 11 partners with the open-source community, many projects of which are used as the base of enterprise technology. Storj is ultimately more centralized than Archon through the use of and reliance on Satellites throughout their iteration of the architecture described in their newest whitepaper.
Archon is currently conducting a Beta Test, has completed a NEO protocol integration, and is currently onboarding users and dapps into their beta. Raising $2.1 MM in a private placement raise led by Neo Global Capital in 2018, Archon has delivered at a much faster rate than other projects in the decentralized storage space. Archon hopes their advantage of crosschain data interoperability, better performance, and easier user experience gives them an edge to become the dominant player in the decentralized storage space. They are well on their way with partnerships such as Neo, Harmony, and Everipedia.
