• R

    Dear Ultrain community members,

    We are excited to announce the launch of Ultrain’s mainnet monitoring page! Click the following link to see the operational status of our permitted mainnet including mining profitability, real-time consensus visualization, and existing mainnet users:

    The addition of this visibility feature will further explain Ultrain’s consensus procedure and business model. Please visit the Mainnet Monitoring service (link provided above) and read the supporting article at These resources assist us to comprehend Ultrain’s token economic system, and understand how Ultrain’s model is based on a notion of servicing and empowering the fundamentals of the real economy.

    Ultrain business development model is dependable and scalable; as the demand for blockchain services increases, Ultrain’s economic system will grow accordingly to meet the interests of early adopters, as well as ensure the healthy development of the entire business ecosystem.



    Since the call went out for Ultrain’s second round of testnet miner recruitment, we received many applications from all over the world. Thank you all once again for the great interest!
    The following is the list of successful applicants – welcome to the Ultrain testnet family and enjoy your preview of Ultrain’s technology.


    (The list below represents the selected miners since December 26 2018 to January 13,2019)


    Ultrain encourages everyone to continue submitting testnet applications, and stay tuned for more updates from the team!

    Thank you all for your continued support of Ultrain.

    Ultrain Team
    2019. January 25th

    posted in Economic read more
  • R


    感谢大家一直以来对于Ultrain的支持!Ultrain正式宣布:商用主网监控页面已上线,各位可通过此页面查看商业主网的运行数据,了解Ultrain的共识规则和商业模式,并可结合此文充分了解Ultrain Token经济体系,明白Ultrain的经济模型建立于服务实体经济的基础之上,是一个更为良性和符合商业发展规律的模型。随着算力购买的商业需求的增长,Ultrain经济体系将会良性发展,保证早期进入者的利益,从而也可以保证整个商业生态的健康发展。





















    posted in 经济模型 read more
  • R

    To all community members of Ultrain:

    Ultrain is pleased to announce that its permitted mainnet is now fully operational and able to service external enterprises.

    Ultrain’s permitted mainnet entered the beta stage on 1 October 2018. In the 86 days since then we have completed 743,248 blocks on the chain and deployed five smart contracts for two DApps. Transactions peaked at 345 TPS over the 86 days of operation, demonstrating the reliability, usability and performance of Ultrain’s network.

    We have made the following updates to our permitted mainnet:

    1. Optimized the consensus mechanism. We have achieved excellent performance, with 3000 TPS and 1000 nodes in the secure environment of Amazon cloud by July 2018. Now we have achieved the innovative optimization of the random algorithm component of our existing consensus mechanism. In the public network environment we are maintaining 1000 TPS (capable of carrying 80 million daily transactions), while improving the stability and usability of network services. As a permitted mainnet providing commercial services to clients, Ultrain commits to delivering performance of more than 1000 TPS, with 99.99% of the main net functionality usable.

    2. Economic model. We have completed the update process for code in the Ultrain economic system (the specific economic system can be seen in the following link: With this upgrade, DApp developers can choose either “dedicated side chain” or “side chain on demand” to purchase Ultrain’s trust computing services. Moreover, Ultrain’s system is compatible with mining rigs, which can now start the mining process.

    Ultrain has completed its network preparation and is able to provide external parties with commercial services, Ultrain have reached strategic cooperation agreements with Unitopia, Mixmarvel Technology, Real Event Limited, Jungo Network, Great Resource New Energy and other leaders in various industries. We expect numerous commercial DApps to make use of Ultrain’s permitted mainnet in the next few months, providing services to end users while operating on Ultrain’s chain.

    A total of 80 mining rigs are providing computing services in the existing permitted mainnet. These rigs are operated mainly by Ultrain, as well as by business partners. As more companies use and purchase the trust computing services of Ultrain, we hope more mining rigs will join the network to provide computing services. Meanwhile, our mining rigs are open for sale. With more miners joining Ultrain’s business ecosystem, together we will build stable and efficient trust computing services.

    In addition, Ultrian’s commercial packaged service is now available online for purchase and for further detail you may refer to the following link at, to experience Ultrian’s testnet service.

    Warm regards for a wonderful holiday season
    Ultrain Team

    posted in Consensus read more
  • R





    1. 优化共识机制:我们在今年7月已经在亚马逊云上1000个节点的环境下达到了3000TPS的优异性能,而本次通过对共识机制中随机算法部分的优化和创新,我们成功地在公开互联网环境下将TPS稳定到1000TPS以上(即可以承载8000万日活的应用),并有效地提升了网络服务的稳定性和可用性,作为提供商业服务的准入制主网,我们承诺Ultrain性能最低1000TPS,主网的服务可用性达到99.99%;
    2. 经济模型机制:本次升级我们已经完成了Ultrain经济体系代码的更新,具体经济体系介绍可见该链接;通过本次升级,DApp开发者已经可以通过选择“专属侧链”或“按需定制侧链”的方式申请购买Ultrain信任计算服务;同时系统也已经具备了接入Ultrain矿机进行挖矿的能力;

    Ultrain准入制主网已经完成了对外商用服务的准备,同时,我们已经与包括电魂区块链实验室,幂玛科技,Real Event limited,君戈网络,宏日新能源等多家行业内知名公司签订了战略合作协议,在未来的数月内将陆续有一系列商业DApp采用Ultrain准入制主网的信任计算服务,落地到Ultrain链上为用户提供服务。



    posted in 共识 read more
  • R

    Dear Ultrain community members,

    Thank you all for the overwhelming interest and support for Ultrain’s recent public testnet application. We have received a large number of applications, with candidates applying from all over the world, many of which have a technical background in blockchain. Others have backgrounds in traditional industries, but have demonstrated extreme motivation and passion for Ultrain.

    We are pleased to see the widespread awareness of our public chain technology growing, and were impressed by the sincerity with which applications were made.

    Ultrain would like to officially announce our first nineteen successful applicants for the testnet! The shortlist below contains participants from all over the world, from places such as China, the United States, South Korea, and Singapore. These early adopters will be providing computational power for the operation of Ultrain’s testnet, and will preview Ultrain’s technology to help complete testing prior to mainnet launch.

    We have already sent an official email to each of the following persons. If you are listed below, please pay attention to your email.

    1. 5766655@***.com
    2. 739050848@***.com
    3. bigthomsan@***.com
    4. yooobuntu@***.com
    5. somtam4455@***.com
    6. spiritearth@***.com
    7. ansqld@***.com
    8. a9876543245@***.com
    9. yang.liu@***.com
    10. shenzhunallen@***.com
    11. shanhaimeng2013@***.com
    13. ucqwertyui@***.com
    14. 785807763@***.com
    15. lauravn0904@***.com
    16. 822057413@***.com
    17. zhouchuang@***.com
    18. ironman_uestc@***.com
    19. tesirocyt@***.com

    Ultrain welcomes everyone to continue submitting testnet applications, and stay tuned for more updates from the team!

    Ultrain Team
    2018. December 26th

    posted in Consensus read more
  • R




    1. 5766655@***.com
    2. 739050848@***.com
    3. bigthomsan@***.com
    4. yooobuntu@***.com
    5. somtam4455@***.com
    6. spiritearth@***.com
    7. ansqld@***.com
    8. a9876543245@***.com
    9. yang.liu@***.com
    10. shenzhunallen@***.com
    11. shanhaimeng2013@***.com
    13. ucqwertyui@***.com
    14. 785807763@***.com
    15. lauravn0904@***.com
    16. 822057413@***.com
    17. zhouchuang@***.com
    18. ironman_uestc@***.com
    19. tesirocyt@***.com


    posted in 共识 read more
  • R

    Public blockchains are often evaluated based on the pioneering features of their technology. Moreover, the most important aspect of a project’s success is often its economic design. A reasonable and logical model can preserve the overall security of a blockchain through economic incentives. The possibility of a 51% attack can be eliminated using game-theory concepts. In addtionm, widespread network involvement should be encouraged by ensuring fairness between participants through a profit-sharing mechanism. Only win–win outcomes can sustain the healthy growth of a blockchain ecosystem; it is not a zero-sum game.

    Let’s take a closer look at two of the economic models that currently exist in the blockchain world: Proof of Work and Delegate Proof of Stake.

    The Proof of Work consensus mechanism is demonstrated by Bitcoin and Ethereum, both of which involve millions of mining rigs. As the number of miners participating in a PoW network grows, challenges emerge with which each key ecosystem stakeholder – miners and developers – must deal.

    PoW miners encounter the following three challenges:

    1)Dilution of computing power. The potential to successfully mine a block with a single mining rig will decline as more miners join the network. In the Bitcoin network, for example, a mining rig is selected and rewarded with 12.5 Bitcoin every 10 minutes. As more participants join the network, however, the profitability of each miner declines as the potential of being selected for reward declines. The aggregated computing power of the Bitcoin network grew by 400% between October 2017 and April 2018, indicating that the return on investment of individual miners in April 2018 is roughly one-quarter of the October 2017 value.

    2)Volatility of token price. The final output of a miner’s effort is a token, and the price of each token directly affects a miner’s investment return. A volatile market will have a direct impact on miners. Of the two existing PoW platforms, Bitcoin aims to stabilize volatility by decreasing the number of Bitcoin rewards by half every 210,000 blocks. Ethereum, on the other hand, issues a fixed number of tokens per block. Both approaches have had only limited success in stabilizing token prices.

    3)Liquidity of mining rigs. Most PoW-based mining rigs are customized ASIC devices specifically designed for solving their particular algorithms. If they are not used for mining, such mining rigs would be worth less than the metals they are built on – they have no true value. For miners, therefore, their mining rigs depreciate at stunning speed after purchase.

    Developers encounter the following two challenges:

    1)Cost. DApp developers must pay high blockchain service fees, regardless of the platform that they are operating on. The usage fee for a DApp with 10,000 daily active users (DAU) can cost around RMB 2 million, which is unaffordable for individual developers and mid-sized development firms. The truth is, apps with only 10,000 DAU are usually seen as trivial, and are usually perceived as a prototype projects in the Internet industry.

    2)Cost management. DApp developers purchase services through tokens, making it extremely difficult to predict a DApp’s system usage fee because of the volatility of token prices. This poses a great challenge for DApp developers.

    The other main blockchain economic model is the DPOS system, made popular by the well-known platform EOS. One of the original intentions of EOS’s model was to reduce costs for DApp developers by not charging a fee for running codes. To avoid the wastage of system resources, however, EOS requires participants to stake EOS tokens for the use of resources such as CPU, RAM and hard-disk space. Such resources are limited, and their price has been affected by speculative purchasing. Currently, on the EOS platform, a DApp with 10,000 DAU needs to stake EOS tokens with a fiat value of RMB 3.5 million, far exceeding the operational budgets of regular DApp development teams.

    Overall, then, significant challenges remain in the econometric design of public blockchains. In our view, the breakthrough lies in the creation of a self-sustaining and self-circulating economic system. At this stage in the blockchain world, it is difficult to create independent ecosystems for individual public chains. Only when the blockchain economy has fully merged with the real economy (e.g. by enabling the real economy to generate more revenue), will stakeholders in the real economy be willing to pay services fees for blockchain technology. Such an integrated relationship will help stimulate the growth and smooth operation of the entire ecosystem.


    Ultrain’s model aims to consolidate all useable and available computing resources by building a trusted computing service similar to the concept of cloud-computing. Corporate or enterprise customers can benefit from trusted computing because it enables them to reconstruct their existing business model, significantly reduce the cost of trust in the business environment, and achieve revenue growth. The fees paid by corporate customers are allocated to all parties in the Ultrain ecosystem to ensure win–win situations for all participants and guaranteeing the continuous development of the entire ecosystem.

    Thus , Ultrain’s economic design involves modelling a new economic system for the blockchain industry with the goal of completely resolving the difficulties faced by the various ecosystem stakeholders – miners, DApp developers and other community participants. We refer to Ultrain’s economic model as A Real Economy Model Based on Blockchain.

    Let’s discuss the design of Ultrain’s economy from two perspectives: allocation, and mining.

    A. Economic allocation:


    Ultrain’s economic model addresses the process by which DApp developers pay usage fees to the Ultrain public chain, and the Ultrain network shares this income with other parties participating in the ecosystem. Revenue will be allocated as follows:

    · Miners. 80% of the total revenue received by Ultrain in service fees will be distributed to the miners in the network. This high proportion is because miners own the mining rigs and therefore contribute resources for the daily operation of the network.

    · Ultrain technology team. 5% of the total revenue received by Ultrain in service fees will be allocated to Ultrain’s technology team. This team is responsible for a range of technology-related tasks, such as the development, maintenance and upgrading of the network. Ultrain’s team will receive a corresponding return for their effort in the project. The reward will be in line with the development of the ecosystem, which motivates members to continue expanding and improving the network.

    · Ultrain community team. 5% of the total revenue received by Ultrain in service fees will be allocated to Ultrain’s community team. The promotion of our public chain technology relies largely on the support and contribution of our community, which is structured around Hubs, with several local hub leaders responsible for operations, marketing and promotion tasks in specific locations. Ultrain will use 5% of total revenue to reward the top-performing Hub teams. Performance will be judged through a voting process executed by community members.

    · DApp teams. 10% of the total revenue received by Ultrain in service fees will be allocated to DApp developers. The prosperity of a public chain is determined by the number of DApps and DAU activities operating on the chain. Ultrain’s revenue generated through public chain service fees will be shared with DApp developers each year or season. However, such profit-sharing will not be apportioned equally among all DApps but, rather, structured to act as an incentive mechanism to reward DApps for excellent performance. Again, the performance of each DApp will be judged through a voting process executed by community members.

    Below, we discuss Ultrain’s design for miners’ investments and their potential returns.


    Ultrain employs a main-chain and side-chain structure. The main chain is used to confirm user information, create new accounts, and transfer assets; it allows smart contracts only for systematic operational purposes and does not allow the deployment of smart contracts designated for DApp usage. Side chains are designed specifically for DApp deployment based on the needs of individual clients. A single DApp client could be fully serviced by an entire side chain, or a single side chain could simultaneously provide services for multiple DApps.

    Miners provide mining rigs for both the main and side chains. To preserve the overall security of side chains, the main chains will actively and randomly move mining rigs from side chains to main chains to reduce the likelihood of malicious attacks by misbehaving miners. The minimum number of mining rigs required to form a side chain will be 40; a maximum limit of 320 rigs per miner will be implemented to ensure that all miners participating in the network obtain a minimum return on their investments.

    Requirements for miners

    Miners must have:

    1. A rig that meets Ultrain’s minimum requirements: 2.1GHz, 8 cores CPU, 32GB storage, and 100MB bandwidth, staking 42,000 UGAS.

    2. A credit rating of 100. The credit system is used as an indication of the trustworthiness of a rig. Participants can join the Ultrian network and await the formation of 252,000 blocks (which takes approximately 1 month). If no malicious behaviour is detected within this period, a credit rating of 100 will be assigned.

    Like other projects in the blockchain industry, Ultrain is trying to resolve the trustworthiness of mining rigs. Bitcoin uses energy consumed as the mining cost, and the hurdle of 51% computing power, and it achieves trust for mining rigs through the application of game theory. Algorand, on the other hand, assumes that >2/3 of nodes are naturally trustworthy but does not explain the basis of this assumption.

    Ultrain believes that there is no free lunch when it comes to trustworthiness and therefore will require mining participants to “pay their dues” to secure the network. In contrast to Bitcoin, however, Ultrain will not use energy resources to determine mining cost; rather, it will have two mechanisms:

    1. Staking 42,000 UGAS and thereby losing the ability to liquidize these assets is in itself a cost.

    2. Participants are required to obtain and maintain a credit rating of 100. Before joining the Ultrain network, miners must wait for the formation of 252,000 blocks (approximately 1 month) without any form of malicious behaviour – this is, in effect, a time cost.

    Together, these two types of cost form the cost of participation for Ultrain mining rigs.

    Ultrain uses a custom consensus mechanism - Random Proof of Stake. RPoS consensus is similar to the PoS mechanism, but with a random selection process. It increases the security of the network by requiring token staking to prevent a 51% attack. If a malicious attack is detected within the network, the token stakes of such participants will be seized automatically and their credit rating reduced to 0; thus, this is the cost of malicious behaviour in the network.

    Mining stage

    There are two phases of Ultrain mining, one for credit-rating establishment and the other for proving the calculation service online.

    Credit establishment phase

    When a new mining rig enters the Ultrain network it enters a queue of mining rigs that are hoping to receive a credit rating after waiting for the formation of 252,000 blocks (approximately 1 month). When a rig has achieved this, it may enter the next phase of providing calculation on a chain. The rigs queueing at this stage are offered no reward or incentive.

    Proving calculation phase

    Mining on main chain

    The main chain is constructed to service the side chains (those parts of the network providing computing services to clients). Thus, the main chain will charge 2% of the total fee. The scheduling of mining rigs in side chains, transfers between accounts, the creation of accounts, and other functions will all require UGAS when executed. All UGAS will be directed into an aggregated account and subsequently allocated, pro-rata, to individual mining rigs, depending on the time each rig stays online on Ultrain’s network. Hours will be added when the rig is continuously online and subtracted if the rig goes offline.

    Mining on side chains

    If DApp developers purchase new computing services but no calculating power is available in the network, Ultrain will assign and select rigs from the main chain to form new side chains and provide additional calculation services.

    Such assignments and selections will be executed through smart contracts via a fully automatic process involving the following steps:

    1. Rigs with high credit scores will have a higher likelihood of selection on the side chain. The minimum credit score for selection as a side-chain rig is 100.

    2. The minimum number of rigs required to form a side chain is 40 rigs. If the side chain is yet to fulfill this requirement, new qualified rigs will be automatically assigned to this side chain.

    3. When multiple side chains require more rigs, qualified rigs will be assigned with equal distribution among side chains, until specific side chains reach the maximum number of allowable rigs.

    Side chains will employ RPoS consensus, with nodes generated at a rate of 1 block every 10 seconds. Such nodes will be rewarded with 1 UGAS; the mining rigs selected for the side-chain computation will not be assigned back to the main chain and will continue providing service on the side chain.

    DApp service purchase

    Ultrain provides two methods of fee calculation for enterprises.

    1. Customized specific side chain. A side chain may be dedicated to service a single client, depending on their specific requirements. The following criteria will help clients evaluate their true needs:

    · TPS requirement: a single Ultrain side chain is capable of providing services with 1,000 TPS. Enterprises can evaluate their needs based on DApp metrics such as daily active users; they may choose to employ a single or several side chains, depending on their usage.

    · Security requirement: side chains are formed by a number of rigs; the more mining rigs operating in a given side chain, the more secure the environment. A minimum requirement of 40 rigs applies, and 80 rigs are the recommended configuration

    · Usage requirement: This is the estimated amount of time an enterprise will need to operate its DApp on Ultrain’s chain. A minimum requirement time of 1 month applies.

    1. Randomly customizable side chain. Enterprise clients can purchase portions of aggregated computing power when deploying their DApps on the chain, and they can share single side chains with other clients. Publicly shared chains, will be composed of 320 rigs. The CPU, storage and hard-disk resources are aggregated into a package that is divided into 1,000 portions that can be purchased. Each portion of such packages is capable of supporting 2,000 transactions per day.

    Enterprise clients can estimate the service fee and stake the corresponding amount of UGAS tokens in a smart contract operating on the main chain. The service can start as soon as the prerequisite computing power has been assigned from the main chain to side chain.

    The amount of tokens staked by an enterprise will be used to generate blocks every 10 seconds, rewarding the node that generated the latest block with 1.25 UGAS.

    For example, if an enterprise client was to construct a single side chain specifically for its own disposal, the amount of UGAS required to be staked on the network would be calculated at 1.25*(606024*365)/10 = 3,942,000 UGAS for one year, since 1 UGAS is allocated every 10 seconds to the rig that completes the latest block and simultaneously generate 0.25 UGAS to Ultrian's system account for future allocation to DApp, community or technical team.

    C. Summary of the advantages of Ultrain’s economic model:

    The model fundamentally resolves the key issues and concerns of DApp developers and miners, as follows:

    For miners:

    1) Dilution of computing power. All mining rigs participating in the Ultrain network can achieve a specific range of return on their investment in the network because an upper limit exists on the number of rigs in any given side chain. Computing power will diminish, but only to a certain point.

    2) Volatility of token price. The tokens DApp developers use to pay Ultrain for its service will be staked and locked on the main chain and then redistributed back to miners at a consistent rate. This mechanism will maintain token price stability.

    3) Liquidity of mining rigs. Most mining rigs operating on the network will likely be low-end devices commonly found on the market. If miners no longer want to mine they can easily trade their rigs in the second-hand market.

    For DApp developers:

    1) High cost. Ultrain provides DApp developers with a package deal that allows portions of the aggregated resources such as CPU, storage, hard-disk space and bandwidth to be used based on the client’s needs. This package deal would only cost RMB 30,000 per year, which is more reasonable and affordable compared with competitors that charge millions.

    2) Cost management. By using the staking mechanism, DApp developers can be sure that the rate of service will be consistent over a given period because the staked tokens are locked in. This significantly reduces cost volatility.

    In conclusion, Ultrain’s economic model is built on the idea of serving the real economy and is therefore both practical and reasonable, as well as capable of continuous expansion as more computing power is added to the network. Its smoothly designed economic system would guarantee the interest of early participants and help ensure the healthy development of the entire ecosystem.

    This write-up constitutes our initial vision of the Ultrain economic model. We know there is room for improvement and hope that the community will participate in the improvement process through Reddit, Telegram and WeChat groups.

    posted in Economic read more
  • R





    1. 算力稀释问题:大家知道,POW挖矿是随着参与矿机越多,单台矿机(或者说单位算力)收益越低的过程,以比特币举例,现在比特币网络每10分钟会有一台矿机被随机选中出块,同时奖励给该矿机12.5个比特币,所以全网矿机数量越多,每台矿机可能被随机选中的机率就越低,每台矿机的收益也就降低了,我们如果看一下17年10月份到18年4月份矿机算力的变化,这个期间算力增长了四倍,相当于矿工在这个期间的投资回报只有原先的1/4;

    2. 币价稳定性问题:矿工的产出是矿机挖到的Token,而Token的价格直接关系到矿工的收益。如果币价产生大幅度的波动,会直接对矿工的收益产生大幅度的影响,现有的POW机制,比特币是通过四年减半的方式,人为的控制比特币的总产量,从而期望实现币价的正向增长;而以太坊则采用了每年增发的方式,更没有考虑到如何保障币价,这两种方式都很难解决币价稳定性的问题;

    3. 矿机流动性问题:由于现在的POW矿机大部分都是定制的专有ASIC矿机,如果不挖矿,这些矿机基本与废铁一样,没有任何价值,所以对矿工而言,这些矿机资产很难保值;


    1. 成本过高的问题:无论是基于以太坊还是EOS进行DApp开发,DApp开发者都需要支付高昂的区块链服务使用费,以一个日活1万的应用举例,部署在以太坊上需要每年支付200万人民币左右的使用费,这对一个中小开发者而言是完全无法承受的,而日活1万的App在互联网上基本可以忽略不计,只能算一个互联网产品的原型阶段。

    2. 成本难以控制的问题:由于DApp开发者是通过Token购买服务,随着Token价格的涨跌,DApp的系统使用费变的极难预估,这对DApp开发者而言也是一个极大的风险;

    除了POW体系外,现在还很流行的一个区块链经济体系就是DPOS体系,其中最著名的代表就是EOS。EOS经济体系设计的初衷之一就是希望降低DApp开发者的成本,不收取代码运行的手续费,而为了避免系统资源被无意义的浪费耗尽的问题,EOS的经济体系是通过抵押EOS Token租用系统的CPU、RAM和DISK等资源,这个时候问题就出现了,由于这些资源的总量是有限的,这些稀缺资源立刻导致了投机客的疯狂抢购,现在在EOS平台上,1个日活1万的DAPP,需要抵押的Token,在现在的币价折合人民币约350万元,这远远不是一个普通的DApp开发团队可以支付的。









    l 矿工:使用费的80%。由于矿工是实际矿机的所有者,提供了硬件设备并维护了网络的正常运行,在收益中应该占较大比重;

    l 技术团队:使用费的5%。技术团队承担了Ultrain软件的开发、维护、升级等一系列技术工作,为避免比特币core team的情况出现,技术团队理应在Ultrain的生态发展中获得收益,而且该收益应该与生态经济大小挂钩;

    l 社区团队:使用费的5%。Ultrain区块链公链技术的推广,很大程度上依赖于社区的支持和贡献,Ultrain在社区的建设上,采用Hub机制,我们希望在全球重点城市建立Ultrain Hub,每个城市都有几名hub leader负责Ultrain在当地的运营、市场和推广工作。我们每年会从Ultrain的使用费中分出5%奖励给当年表现优异的头部Hub团队,而对于优异的定义,将会采用社区投票的形式由社区来决定;

    l DApp团队:使用费的10%。Ultrain区块链公链的繁荣,取决于链上DApp数量的多少,以及该DApp的活跃度。我们每年或每季度会从Ultrain的使用费中分出10%奖励给链上的DApp应用,原则上该奖励不会是均分的模式,而是奖励表现优异的DApp,其具体标准将会采用社区投票的形式由社区来决定;








    1. 一台符合Ultrain要求的低端小型机(2.1GHz 8核CPU,32G内存,100M带宽),抵押42000UGas。



















    1. 可信度越高的机器,越优先被调度到侧链提供服务,进入调度队列的最低信用度为100;

    2. 侧链建立的条件,是满足最低侧链台数,即40台;如果该侧链需要的矿机数量不只40台,在有新的符合要求的矿机进入网络后,会继续分配到该侧链;

    3. 多条侧链情况下,新进入的矿机分配到侧链时,采用平均分配法,保证在侧链达到最大矿机台数前,各条侧链矿机数量均匀分配。






    l TPS要求:Ultrain一条侧链提供的TPS为1000,企业可以按自己DApp应用的用户日活量,选择建立一条或多条侧链来提供算力服务;

    l 安全性要求:需要几台矿机组成侧链,台数越多安全性保障越高,最低数量为40台,建议80台;

    l 使用时长:预计DApp使用Ultrain服务的时间,最低1个月;




    比如一家企业希望使用300台矿机的算力一年的算力构建一条专属侧链,(606024365)/101.25 = 3942000个UGas,这家企业需要锁仓这么多的UGas在主链上,并在一年时间内,每10秒分配给300台矿机中的随机一台1枚UGas,并同时分配给Ultrain系统账号0.25枚UGas用于后续分配给DApp,社区及技术团队。




    1. 算力稀释问题:进入侧链的矿机,通过规定侧链机器的最大数量,保障了侧链单台矿机的最低收益,所以算力会稀释,但有一个稀释到的最低值,不会无限的稀释下去;

    2. 币价稳定性问题:DApp开发者购买算力的Token会直接锁定到主链上,并逐步释放给提供算力的矿机,通过这种方式可以有效的稳定Token的价格;

    3. 矿机流动性问题:矿机就是普通的低端小型机,如果矿工不想挖矿,可以很容易的在二手服务器市场出售,具有极佳的流动性。


    1. 成本过高的问题:首先Ultrain提供了打包的计费模式,将CPU、内存、硬盘和带宽统一打包出售,便于计算;其次,在Ultrain上部署一个日活1万的应用,选用随需定制侧链的方式,只需要花费3万人民币/年,远远低于同类竞争对手的几百万一年的使用费。

    2. 成本难以控制的问题:由于采用抵押的模式,DApp开发者可以保证在抵押的这一段时间内,DApp的算力使用成本都是保持一致的,不会出现成本难以控制的情况。




    posted in 经济模型 read more
  • R




    大家知道,我们在今年7月15日发布了一个部署在亚马逊云端的1000个节点的Ultrain概念验证网络,并取得了在完全去中心化的架构下TPS 3000的优异性能成绩。但是,亚马逊云毕竟是一个条件优异的网络运行环境,比如其跨洲网络间的网络传输只有100ms左右的延迟。Ultrain致力于建立一个完全的去中心化的网络,我们希望我们的矿机可以部署在各个家庭之中,既可以做为PC正常使用,闲时又可以做为Ultrain矿机提供算力,实现分布式CPU算力的共享。为实现这个目标,Ultrain需要能在复杂的公开互联网基础上良好运行,我们需要解决包括洲与洲之间的巨大的网络延迟;高TPS下巨大的网络传输流量导致对网络流量的高要求;各个国家网络基础设施的差异带来的特殊处理(比如针对中国网络环境下,没有公网IP的机器之间如何对用户透明的相互访问与通信);互联网网络拓扑结构不可控带来的网络风暴的风险等问题。这其中的许多问题,都是其他公链从没有尝试去解决的,而我们Ultrain希望能真正的在区块链领域解决这些问题,从而实现真正的去中心化的区块链公链。







    posted in 共识 read more