从技术创新角度详解,Aptos凭什么被称为杀手级新生态?

老雅痞Published on 2022-12-28Last updated on 2022-12-28

Abstract

作为明星公链,Aptos一经推出被称为SOL杀手,其优秀之处备受不少项目方的喜爱。

作为明星公链,Aptos一经推出被称为SOL杀手,其优秀之处备受不少项目方的喜爱。

Aptos通过更好的技术和用户体验为Web3带来了范式转变。作为市场上延迟最低,吞吐量最高的区块链,它使开发人员能够高效地构建并创建最佳应用程序。Aptos社区使Aptos得以蓬勃发展。除了社区,强大的技术是其另一个超能力。

在今天的内容中,我们整理了推动Aptos重新定义Web3以实现更具意义的用途的创新。这些创新协同工作,为开发人员和用户提供了无缝体验。无论你是使用Move语言轻松开发,还是受益于技术支持的用户保障,Aptos都提供了流畅的体验。

真诚建议各位builder一起来了解一下。

1、用于更好构建的Move语言

Aptos上的Move为可扩展的、面向未来的应用程序创造了更好的开发者体验。Move和Aptos框架一起,通过有意消除通常限制开发人员可访问性的不必要的复杂性,提高了建设者的效率。

Move是一种强调安全并专为安全资产管理而设计的表现力强且易于使用的编程语言。Move为Web3带来了Rust为基础设施开发带来的东西——一种安全、快速、富有表现力的映射交互方法。

Aptos区块链原生集成了Move语言。Aptos和Move共享许多相同的核心设计原则,使Aptos成为了高效、愉快的Move开发之家。Move最初是为Aptos区块链的前身设计的。这种关系确保了现有的Move开发人员可以无缝地在Aptos上进行构建,而新的Move开发人员可以从Aptos之前的文档、指南和示例中受益。

Aptos通过在语言和框架层面增加了完善的安全架构、详细和可配置的gas计量、代码可升级性、资源帐户等多种功能,极大地改善了Move生态系统。除此之外,Move Prover为合约不变量提供了额外的保障,并在Aptos中得到了积极扩展。

许多Move语言的原始研究人员和开发人员继续在Aptos生态系统中构建,从而发展了该语言和Move社区。通过四年的测试和认证,Move语言证明了它安全且可用于生产。

2、带来更多编程自由的Block-STM

Block-STM是一种新的智能合约并行执行引擎,它围绕着Aptos的软件交易内存和optimistic并发控制原则构建。这种新的交易并行方法可以在不影响开发人员体验的情况下实现更快的交易处理。

与要求预先了解要读/写的数据而打破交易原子性的并行执行引擎不同,Block-STM使开发者能够无限制地编码,并为实际用例实现更高的吞吐量和更低的延迟。

开发人员可以使用Block-STM轻松构建高度并行化的应用程序。与通常需要将操作拆分为多个交易(破坏逻辑原子性)的其他并行执行环境相比,Block STM支持更丰富的原子性。这通过降低延迟和提高成本效率增强了用户体验。

3、链上治理和去中心化

为了支持真正去中心化和无许可的第一层,Aptos内置了链上治理,实现了网络和虚拟机配置的无缝更改。Aptos测试网3和主网证明了这一点。

在主网上,这通过降低“投票权增加限制”提高了网络的可靠性。设置更激进的阈值可以快速引导网络。超过52%的代币所有者投票支持这一提案,以帮助保护网络安全。

4、AptosBFTv4效率共识

AptosBFTv4是第一个具有严格正确性证明的生产型区块链BFT协议。该协议可以提供低延迟和高吞吐量,并充分利用底层网络的优势。Aptosv4BFT在Hotstuff的基础上进行了改进,将提交延迟从3步减少到2步,在不牺牲通信复杂性的情况下将延迟降低了33%。

AptosBFTv4是流水线式的,类似于处理器,以确保在每个步骤中最大限度地利用资源。因此,单个节点可以参与共识的许多方面,包括选择将哪些交易包括在一个区块中,执行另一组交易,将另一组交易的输出写入存储,以及验证另一组交易的输出。这使得吞吐量仅受最慢阶段的限制。

在实施过程中,考虑到安全性和可升级性,该实施方案干净利落地分离了用于隔离的不变量和强制执行不分叉思维的有效审计。同样的软件堆栈已经经过了4次升级,并在实时网络上进行了测试,证明了其开发过程的周到和稳健。在其第四次迭代中,AptosBFTv4是最快的、可用于生产的拜占庭容错共识协议。

Aptos确保了整个网络正常运行时间,即使在个人停机期间也是如此。这由一个链上信誉系统维护。该信誉系统将过去的可用性和性能作为未来的指标,自动将响应迟钝和表现不佳的验证者的负面影响降到了最低。

5、增强消费者信心的保障措施

Aptos帐户支持灵活的密钥管理,其中包括对密钥轮换,加密敏捷性和混合托管模式等功能的支持。密钥轮换通常对于防止可能危及多方账户的远程攻击非常重要。在其他区块链上,轮换只能通过将所有资产迁移到新帐户来实现。帐户与密钥解耦的方法使Aptos能够无缝地添加新的数字签名算法,以支持公钥和私钥类型。混合托管模式支持高级恢复解决方案和帐户管理,以帮助弥合Web2和Web3之间的差距。

钱包可以使用交易预执行,在用户签署交易之前向用户解释交易的结果。在签名前评估交易可以降低网络钓鱼攻击等在Web3中变得越来越普遍的安全风险。为了进一步强化用户体验,Aptos区块链限制了每笔交易的可行性,并通过三方面的保护(序列号、过期时间和链ID)保护签署人免受无限有效性的影响,以防止错误/攻击。

Aptos共识协议和经过认证的存储为轻客户端协议提供了无缝和实际的支持,使用户体验更安全、更值得信赖。Aptos网络公开欢迎任何人连接一个全节点,以获得对认证数据的直接访问。这强调了Web3的口号“不要信任,要验证”。为此,Aptos构建了一个高效的多播树结构,提供了一个高吞吐量、低延迟的网络,向参与者传播区块链状态。参与者可以处理自genesis以来的所有交易,也可以完全跳过区块链历史,只使用路径点同步最新的区块链状态。轻客户端可以同步特定帐户或数据值等部分区块链状态,并启用验证状态读取,例如使用BFT时间戳获取验证帐户余额。

6、面向未来的网络模块化架构

从一开始,Aptos系统内的每个区域就是以模块化和灵活性的心态进行设计的。这使得Aptos架构能够支持频繁的升级,这反过来意味着区块链可以快速采用最新的技术进步,并为新的和正在出现的用例提供一流的支持。

Aptos模块化架构设计为客户端提供了灵活性,并为频繁升级进行了优化,实现了零停机时间——这些功能在以前的主网迭代、测试网和许多内部压力测试中都得到了验证。Aptos区块链包括了嵌入式链上变更管理协议,以快速部署新技术创新并支持新的Web3用例。

7、基于提议的奖励系统

在Aptos Incentivized Testnet 3中,我们利用基于投票的奖励系统对节点运营商进行了奖励。在该模型中,一旦三分之二的投票到达提议者节点,就实现了BFT下的共识。这意味着三分之一的后期投票不包括在内,其相关的验证者也不会得到奖励。

这可能会导致延迟竞争,靠近主节点集群的验证者往往会获得更多奖励。在这种情况下,节点运营商会把他们的节点移到离主集群更近的地方,以改善延迟,从而增加他们的质押奖励。这损害了去中心化,因为它激励了协同定位。

我们现在已经根据提案的表现实施了奖励,以促进更大的去中心化。

基于提议的系统比投票有更高的timeout,可以说对跨区域的延迟不那么敏感。这提高了位于较偏远地区的节点的奖励率,并抑制了如果验证者不在最大的节点集群中,他们的奖励就会减少的地理分布的影响。我们的奖励模型将继续定义投票行为,因为良好的投票表现会影响提议者当选的概率。

8、高性能稀疏默克尔树

Aptos使用Jellyfish Merkle Tree (JMT)设计,该设计利用单调递增的基于版本的密钥模式来优化基于LSM树(或日志结构的合并树)的底层存储引擎(如RocksDB)上的写入扩展。JMT在CPU、I/O和存储占用之间达到了一个实用的平衡点,确保了令人满意的性能,而不需要处理磁盘上庞大且难以处理的膨胀状态数据。

除了将JMT作为Aptos状态的持久化格式外,它还有另一个内存中的无锁稀疏Merkle树实现。它专门为缓存和并行化量身定制,与Block-STM配合使用,以促进高性能的全局状态更新。

Related Reads

Claude Requires ID Verification and Facial Recognition? The Facial Recognition Requirement is an Old Story from Two Months Ago, and "Sharing Data with Police" is a Misinterpretation

Anthropic's updated privacy policy, effective July 8th, has sparked misinterpretations in Chinese social media, primarily concerning new identity verification and data sharing with law enforcement. A detailed comparison reveals these claims are largely unfounded. First, identity verification (including submitting government ID and a live selfie via third-party provider Persona) is not a new July policy. This mechanism was actually implemented in mid-April 2026 for certain high-use or flagged accounts, particularly Claude Max subscribers. The July update merely formally documents this existing practice in the policy text under a new "Verification Data" section. Second, the widespread claim that the new policy lowers the threshold for sharing user data with law enforcement is incorrect. Comparing the new text with the old version (dated September 28, 2025) shows no substantive tightening. While the new policy more clearly structures the conditions for disclosure—including having a "good-faith belief" it's necessary for legal compliance, preventing harm, fraud detection, or enforcing terms—the old policy already allowed Anthropic to disclose data based on its judgment for similar reasons (e.g., protecting safety, preventing fraud, or complying with law). The term "good-faith belief" acts as a limiting standard, not a lowered barrier. A 2025 court case where Anthropic resisted disclosing user data in a copyright lawsuit further demonstrates the complexity of such standards. The policy's actual substantial changes address data flows for Claude's Agent capabilities. New clauses clarify that when users connect third-party services or instruct Claude to perform multi-step tasks (reading files, sending messages), their inputs, outputs, and instructions are shared with those third parties, governed by the third parties' own policies. This update fills a compliance gap for Claude's evolving functionality beyond simple Q&A. Other additions include a "Research Participation Data" section and refined marketing legal bases. Anthropic reaffirms core commitments: not selling user data, keeping Claude ad-free, and allowing users to control if chats are used for model training. Overall, this update is primarily a compliance catch-up to existing product features, not a significant new privacy tightening. The heightened concern stems from conflating April's verification rollout, standard legal clauses, and the genuine new provisions regarding Agent tasks.

marsbit2m ago

Claude Requires ID Verification and Facial Recognition? The Facial Recognition Requirement is an Old Story from Two Months Ago, and "Sharing Data with Police" is a Misinterpretation

marsbit2m ago

The World Cup Has Only Just Begun, But AI Predictions Already Have Models Hailed as 'Godly' and Others Flipping Over

After only a few days of the World Cup, AI models are being widely used for match predictions, with mixed early results. These models analyze details like scores, upsets, red cards, and key players, offering users in prediction markets an extra layer of analysis beyond odds and news. Qwen gained early attention for its remarkably accurate calls on the opening day, correctly predicting Mexico's 2-0 win over South Africa and Korea's 2-1 victory over the Czech Republic, while also highlighting red card risks and match flow. Copilot had its own highlights, accurately forecasting the Mexico 2-0 result, the Korea 2-1 win, and a surprising 1-1 draw between Brazil and Morocco. However, it also misjudged several matches, like predicting a Swiss win that ended in a draw with Qatar and missing Australia's upset over Turkey. ChatGPT provided detailed pre-match analysis and correctly called the Mexico 2-0 score, explaining factors like home-field advantage. Yet, it struggled to anticipate upsets, often siding with the stronger team on paper, as seen in its missed calls for the Australia-Turkey and Japan-Netherlands matches. Social media tests pitted models like Gemini, Grok, and Claude against each other for the same games, revealing different predictive "scripts" even for the same fixture. Overall, while AI models like Qwen and Copilot have shown promising, high-profile successes in early matches, their consistency and ability to predict genuine upsets remain in question. As the tournament progresses, more data will be needed to determine which models offer the most reliable insights for prediction markets.

Odaily星球日报6m ago

The World Cup Has Only Just Begun, But AI Predictions Already Have Models Hailed as 'Godly' and Others Flipping Over

Odaily星球日报6m ago

The Unfinished Tale of Jueying, DaXiao Robotics Swiftly "Raises Funds"

Following a major fundraising round involving several prominent investment institutions, DaXiao Robotics, a company backed by SenseTime, has secured hundreds of millions of US dollars in financing for the first half of 2026. This move signals SenseTime's renewed and substantial bet on "Physical AI" through embodied intelligence, following the relative underperformance of its autonomous driving unit, Jueying. While Jueying achieved mass production partnerships in the smart vehicle sector, it failed to become a pivotal player in the high-level autonomous driving landscape, leading to its gradual independence from SenseTime's core financials. DaXiao Robotics now emerges as SenseTime's next major venture into the physical world. The new funding will focus on developing a "world model" and integrated hardware-software solutions for commercial applications like retail, security, and hospitality. This ambition is significantly more complex and capital-intensive than previous projects. A world model requires understanding spatial relationships, physics, and causality to guide robots in long-term tasks, demanding immense computational resources, data, and engineering. The article highlights several challenges. First, the massive funding, while substantial, may still be strained by the high costs of R&D, data collection, and commercial deployment. Second, SenseTime itself, despite narrowing losses, continues its high-investment growth model and cannot solely bankroll this new, expensive endeavor. Third, DaXiao Robotics, led by SenseTime co-founder Wang Xiaogang, carries the technical heritage and resources of its parent company but also potentially its organizational inertia. It operates in a field increasingly dominated by agile, young technical founders. Ultimately, DaXiao Robotics represents SenseTime's attempt to secure a leading industrial position in embodied intelligence—a goal its Jueying unit did not fully achieve in autonomous driving. The new venture starts with strong capital backing, but faces the critical task of rapidly transitioning from technological narrative to sustainable commercial delivery in an early-stage, costly, and highly competitive arena.

marsbit15m ago

The Unfinished Tale of Jueying, DaXiao Robotics Swiftly "Raises Funds"

marsbit15m ago

Crypto 2029: The Ultimate Forecast for the Four-Year Cycle of the Cryptocurrency Industry

Title: Crypto 2029: The Ultimate Four-Year Cycle Prediction for the Encryption Industry This article outlines a detailed, stage-by-stage prediction for the crypto industry from the present to 2029, focusing on tangible shifts rather than abstract theory. Key predictions include: **2026 Mid-Year:** The market shifts focus from traditional tokens to synthetic perpetual contracts for private company shares (e.g., SpaceX on Hyperliquid), which become primary price discovery tools for pre-IPO assets. Most altcoins languish as the market seeks assets with real underlying value. **2026 Year-End:** The "AI + crypto" narrative fades as the AI industry itself does not require crypto infrastructure, except for prediction markets betting on model performance. Concurrently, a quiet institutional adoption of asset tokenization (e.g., money market funds) begins under new regulations like the CLARITY Act, creating a dual economy. **2027:** Major public blockchain foundations pivot decisively to serve institutional clients with compliance tools and enterprise sales, while quietly building infrastructure for a future wave of accredited retail investors. Three sectors hit growth ceilings: private perpetual contracts (due to legal restrictions on marketing), stablecoins (due to political uncertainty ahead of the 2028 US election), and tokenized assets (due to cautious institutional scaling). **2028:** Speculative trading diminishes as market efficiency drains liquidity. A major liquidation cascade in synthetic perpetual contracts exposes the flaw of lacking a legally enforceable underlying asset. In response, regulations are revised to allow marketing of private security secondary sales to accredited investors. This creates a legal, direct market for private company equity, absorbing much of the demand previously met by synthetic derivatives. **2029:** A new bull market emerges, driven not by tokens but by tradable equity in innovative private companies (biotech, robotics, AI). Tokens without legally enforceable claims to real assets lose all liquidity. Successful blockchains become invisible settlement infrastructure. Stablecoins grow steadily at a policy-capped rate. Speculation becomes a niche. Core Questions Answered: 1. **Token Value:** Determined solely by legally enforceable claims to real-world assets. 2. **Tech Adoption:** Achieved through blockchain-based primary/secondary markets for private equity, not through forcing tokens onto tech firms. 3. **Crypto as Infrastructure:** The transition happens silently; the technology becomes a mundane, unseen utility like traditional settlement systems. The entire thesis hinges on one testable variable: by late 2028, whether accredited retail investors gain legal, direct access to private asset markets. If not, the core premise—that legal frameworks, not technology, are the main bottleneck—fails.

Foresight News23m ago

Crypto 2029: The Ultimate Forecast for the Four-Year Cycle of the Cryptocurrency Industry

Foresight News23m ago

Trading

Spot
Futures

Hot Articles

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of SOL (SOL) are presented below.

活动图片