A Solid Foundation for IOTA 2.0

IOTA Foundation ·

Unpacking a Trilogy of Pivotal Research Papers

TL;DR:
New peer-reviewed papers on our approach to the ledger, networking, and consensus form the theoretical backbone of IOTA 2.0, our forthcoming protocol upgrade. The Ledger Paper introduces a reality-based approach to transactions, the Networking Paper offers a 'pre-consensus' through the IOTA Congestion Control Algorithm (ICCA) for efficient throughput and congestion management, and the Consensus Paper ensures informed decision-making using the 'heaviest DAG' approach. Combined, they provide a system that enables efficient conflict resolution, and a fair and dynamic environment for the future of DLT.

The primary focus of the IOTA Foundation’s research department is to design the future of the protocol, develop cutting-edge technology, and fulfill the IOTA vision. Therefore, the development of IOTA 2.0, our forthcoming protocol update, is tightly linked with our research progress.

Our researchers and developers have meticulously rebuilt the entire architecture of IOTA from scratch, emerging as a resolute and robust IOTA 2.0. One way of assessing the quality of this work is by evaluating the quality of our academic output, which in turn serves as a validation of the protocol. Specifically, IOTA 2.0 stands on the shoulders of three scientific papers:

  • The Networking Paper, "Access Control for Distributed Ledgers in the Internet of Things: A Networking Approach", published in IEEE Internet of Things Journal (also available on arXiv).
  • The Ledger Paper, “Reality-Based UTXO Ledger”, published in ACM Distributed Ledger Technologies: Research and Practice (also available on arXiv).
  • The Consensus Paper, “A Leaderless Nakamoto Consensus on IOTA 2.0", published in IEEE Access (also available on arXiv).

With the recent acceptance of the Ledger paper in the journal ACM Distributed Ledger Technologies, all three papers have undergone stringent peer review, ensuring that the proposed solutions are robust and scientifically sound, affirming the quality and innovation behind IOTA’s new design.

In this blog post, we unpack “The Trilogy”, summarising the academic pillars on top of which IOTA 2.0 is built. But first, let’s comment on the interplay of these three papers.

The big picture

IOTA's architecture is a testament to the synergy achieved when multiple research components come together. The Networking Paper's pre-consensus mechanism lays the groundwork, ensuring efficient throughput, and is the foundation for the Consensus Paper. The Consensus Paper is also built on the principles and strategies of a reality-based ledger introduced in the Ledger Paper. Together, these research pieces form a comprehensive framework, demonstrating the advanced capabilities and vision of IOTA 2.0.

Bridging the Networking and Consensus papers

The Networking Paper explains the IOTA Congestion Control Algorithm (ICCA), which acts as a pre-consensus checkpoint. This ensures that, before transactions progress to the consensus stage, they're first vetted for authenticity, weeding out potential malicious inputs. This preliminary evaluation is vital for maintaining the integrity of a decentralized system like IOTA 2.0, ensuring a clean slate of transactions for the subsequent phase.

Once vetted by the ICCA, these transactions are directed towards the main consensus process, as elaborated in the Consensus Paper. Given that these transactions are pre-filtered, the consensus stage can now focus on validation and verification without being bogged down by questionable inputs.

Together, these two papers illustrate IOTA's layered approach: first filtering for integrity with the Networking Paper, then validating with the Consensus Paper, ensuring a seamless, efficient, and secure transaction process.

Bridging the Ledger and Consensus papers

The Ledger and Consensus papers serve as pivotal components within IOTA 2.0's framework.

The Ledger Paper's introduction of multiple realities forms the foundation for the 'heaviest DAG' approach in the Consensus Paper. This integration ensures the consensus mechanism operates with a comprehensive snapshot, making decisions with a holistic view of all possible ledger states or realities.

Leveraging the ‘heaviest DAG’ principle, conflicts among the realities set by the Ledger Paper can be resolved efficiently, eliminating the need for abrupt or complete system halts.

Unpacking The Trilogy

1. The Networking Paper: "Access Control for Distributed Ledgers in the Internet of Things: A Networking Approach"

Diving deep into the infrastructure of IOTA, the Networking Paper presents a framework for how IOTA maintains its decentralized ethos while managing challenges tied to its open-access model. Let's go through its core facets:

  1. Efficient Congestion Control: IOTA 2.0, through the IOTA Congestion Control Algorithm (ICCA), champions high efficiency in regulating throughput. The mechanism serves as a ‘pre-consensus’, enabling nodes to sieve blocks from malicious users. With ICCA, there's an assurance of fairness in both throughput (aligned with the user’s Mana) and latency.
  2. Resource-Aware Mechanism: Recognizing that resources, whether bandwidth or node capabilities, are finite, each block is given a 'work score'. This score captures the amount of resources required from nodes to process that block and is affected by block size and signature, among other factors.
  3. Decentralization at its Peak: The system adopts a deterministic scheduling process, computed in work score per second. This ensures that nodes meeting the minimum hardware criteria can effectively manage and process all visible blocks without being swamped by high-speed devices.
  4. Lightweight Scheduler Protecting the P2P Network: We adopt an efficient scheduling algorithm based on Deficit Round Robin (DRR) to deal with high Mana variance and bursty traffic and to protect the node’s neighbors from spam attacks. The node’s inbox is logically split into several queues, each one corresponding to a different user. The scheduler iterates all queues in sequence and increments a counter (deficit) according to a value proportional to the user’s Mana: if the deficit is greater than the work score of the block at the head of the queue, this block will be scheduled and the value of the counter is decremented by the block's work score.
  5. Local Traffic as a Mirror of Global Congestion: In distributed ledgers all traffic traverses through every node. Hence, local congestion at a node indicates congestion elsewhere in the network. This observation is crucial, as it presents an opportunity for an access control algorithm based entirely on local traffic and without the need for additional, potentially corruptible, messages between nodes.
  6. Permissionless Access: In IOTA 2.0, each user can produce blocks and add them to the ledger without the need for intermediaries. Hence, to regulate the block generation per user in a decentralized manner, the Networking Paper suggests using a rate setter based on the AIMD algorithm, adapting the issue rate depending on local information, that is the issuer queue length. However, since this approach is more suitable for a high-congestion network, we propose a deficit-based rate setter for the launch of IOTA 2.0, which adapts the rate based on the accumulated deficit.
  7. Built-in Resistance to Spam: In the event of congestion, either due to spamming or high traffic, IOTA’s protocol is designed to selectively drop blocks. When the inbox buffer becomes full, the protocol drops a block from the queue with the greatest work divided by the user’s Mana. The paper demonstrates that these blocks, especially those created by malicious users, are dropped without hampering the throughput of genuine users.

Note: The final version of our congestion control also requires issuers to burn a certain amount of Mana to issue blocks. In this way, users will be required to use the valuable blockspace with caution.

2. The Ledger Paper: “Reality-Based UTXO Ledger”

The Ledger Paper, a cornerstone of IOTA 2.0’s foundation, introduces a significant departure from traditional blockchain architecture: an optimistic treatment of incoming transactions before an eventual consensus is achieved. Here’s what this means, broken down:

  1. Reality-Based Ledger: At the heart of this paper is the introduction of a 'reality-based' ledger. In simple terms, this ledger doesn’t just record a single version of events (or ‘reality’). Instead, it allows for multiple, conflicting versions of events to coexist temporarily until a consensus is achieved. This is a radical idea, as it moves away from the classic 'one chain of blocks' model that most are familiar with.
  2. Fast and Parallelizable Ledger Progression: In conventional blockchains, transactions are grouped into blocks that are added sequentially to the chain. IOTA 2.0’s approach allows transactions to be processed in parallel. This means the system can handle multiple transactions at the same time, significantly speeding up the overall throughput and responsiveness of the network.
  3. Optimistic Execution of Transactions: Traditionally, a blockchain must resolve conflicts (two transactions trying to spend the same funds, for example) before adding transactions to the chain. IOTA’s reality-based ledger takes a more optimistic approach: it allows transactions – whether they are independent, concurrent, or even conflicting – to be executed and charted. The network later resolves conflicts and reaches consensus, ensuring that the ‘real’ version of events is agreed upon. This means that IOTA can keep operating smoothly and quickly, even if not all users are in agreement about what has happened.
  4. Optimistic Processing as a Mitigation Strategy: By optimistically processing transactions – meaning the system proceeds with transactions even when there is contention over an output – the network can continue to function smoothly while conflicts are resolved. We see this as a viable strategy to mitigate the contention problem inherent in the UTXO model. In the ‘unpacking’ of the Consensus paper below, we’ll see that this feature develops its full strength in combination with our Conflict resolution mechanism.

This approach represents a substantial leap in the quest for a more scalable, efficient, and robust distributed ledger technology. It’s one of the standout features that sets IOTA 2.0 apart from the pack.

3. The Consensus Paper: “A Leaderless Nakamoto Consensus on IOTA 2.0”

Traditional blockchains have a simple rule for resolving conflicts: the longest chain wins. This is a straightforward way to ensure that all participants in a network agree on the history of transactions. However, IOTA’s unique architecture requires a more sophisticated approach. This is where the concept of the 'heaviest Directed Acyclic Graph (DAG)' comes into play, in which the Tangle itself is used for conflict resolution. Let’s break this down into the main points:

  1. Heaviest DAG: Instead of simply following the longest chain of blocks, IOTA nodes look at a more complex structure: a DAG, known as the Tangle. In this system, the 'heaviest' DAG (meaning the one with the largest number of non-conflicting transactions) is deemed the authoritative version of the transaction history.
  2. Leaderless Consensus Mechanism: Traditional blockchains often rely on some form of leadership: a miner who finds a new block, or a group of validators who agree on the next block. IOTA’s consensus is leaderless, meaning that no single participant has authority over the others. This approach is more democratic and reduces the risk of centralization, enhancing the network's security and resilience.
  3. Streamlining Conflict Resolution in IOTA: In most blockchain systems, each block is a collection of several transactions bundled together. IOTA takes a radically different approach: each block contains just a single transaction. This seemingly simple change has profound implications for how the network operates and resolves conflicts. With each block containing only one transaction, conflicts can be identified and addressed at a granular level.
  4. No Need for Total Ordering: Traditional blockchains must establish a total order of transactions, which means every node in the network has to agree on the exact sequence of all transactions. IOTA’s approach bypasses this need. With single-transaction blocks, the network can resolve conflicts independently and in parallel, without having to agree on the order of unrelated transactions.
  5. Enhanced Network Responsiveness: This structure allows IOTA to operate with a high degree of responsiveness. Because the network is resolving conflicts continuously and independently, transactions can be confirmed more swiftly, allowing the ledger to progress without unnecessary delays.

In conclusion, the Ledger, Networking, and Consensus papers serve as the three cornerstones of IOTA 2.0. Together, they form a comprehensive framework, designed to address the challenges and capitalize on the opportunities of a decentralized future.

Anyone following our progress closely is likely aware that we're currently implementing the principles and methods described above. Coming up next is a full introduction to the IOTA 2.0 protocol, going into the deepest levels of all its aspects, including full documentation: the first blog post in this series outlines the vision behind IOTA 2.0, Digital Autonomy for Everyone. In the meantime, if you have questions or suggestions, feel free to get in touch on our Discord.