Ethereum: Node Types, Mempool & Finality Explained

Key Takeaways

• Ethereum finality takes approximately 15 minutes under normal conditions, requiring two-thirds of validators to attest to checkpoint blocks before transactions become irreversible
• The mempool serves as a temporary waiting area where transactions compete for inclusion based on gas fees, not a single universal pool but rather individual mempools maintained by each node
• Three main node types exist: full nodes (store recent 128 blocks), archive nodes (store complete blockchain history), and light nodes (store only block headers)
• Mempool congestion directly impacts finality timing—higher transaction volumes increase block proposal latency and validator processing demands
• Understanding these systems helps miners optimize transaction timing, manage gas fees, and avoid MEV exploitation risks

Article Summary

Ethereum’s mempool acts as a staging area for pending transactions before block inclusion, while finality represents the point when transactions become permanently irreversible through validator consensus¹. Together with different node types serving distinct network functions, these systems form the foundation of Ethereum’s transaction processing architecture².

What Is Ethereum Finality and How Does Proof-of-Stake Achieve It?

Ethereum finality represents the guarantee that a block and its transactions cannot be altered or removed from the blockchain without burning at least 33% of the total staked ETH³. This crypto-economic security model makes attacking the network financially impractical for any rational actor. Under Ethereum’s proof-of-stake consensus mechanism implemented during The Merge in 2022, finality occurs through a checkpoint system rather than probabilistic confirmation⁴.

Finality provides the ultimate security guarantee—once a block is finalized, it is permanently part of the canonical chain, unlike proof-of-work systems where deep reorganizations remain theoretically possible even after many confirmations.

The Checkpoint-Based Finality System

Ethereum divides time into epochs, with each epoch containing 32 slots that last exactly 12 seconds each⁵. The first block in each epoch serves as a checkpoint. Validators vote on pairs of checkpoints they consider valid through a process called attestation. When at least two-thirds of the total staked ETH attests to a checkpoint, that checkpoint and all its ancestors become finalized⁶. This typically takes 64-95 slots, or approximately 15 minutes under normal network conditions⁷.

The two-thirds threshold creates powerful economic incentives. An attacker attempting to revert a finalized block would need to control more than one-third of all staked ETH and be willing to lose that stake through slashing⁸. As of December 2025, with approximately 35-36 million ETH staked representing nearly 30% of total supply, this represents billions of dollars at risk⁹. This economic barrier makes finality attacks prohibitively expensive and protects the network’s integrity.

How Validators Enable Finality

Validators play three critical roles in achieving finality¹⁰. Block proposers create new blocks by packaging pending transactions from the mempool. Attesters verify proposed blocks by voting on their validity—these votes determine which blocks progress toward finality. Aggregators combine individual attestations into efficient bundles that reduce network bandwidth requirements.

Running a validator requires staking 32 ETH as collateral¹¹. This stake can be slashed—permanently destroyed—if the validator behaves maliciously or goes offline during critical periods. Common slashing events include proposing conflicting blocks or submitting contradictory attestations¹². Minor penalties occur for simple downtime, while major slashing can destroy up to 100% of a validator’s stake during coordinated attacks¹³.

Understanding Ethereum’s Three Node Types

Ethereum nodes come in three distinct types, each serving different purposes and requiring varying levels of resources¹⁴. These nodes form the backbone of Ethereum’s decentralized network, processing transactions, storing blockchain data, and enabling validators to reach consensus.

Full Nodes: The Network’s Backbone

Full nodes store and maintain recent block data while participating actively in block validation¹⁵. They verify all blocks and states, serving blockchain data upon request to support the network. After Ethereum’s transition to proof-of-stake, full nodes require two separate client software components¹⁶. The execution client (like Geth, Nethermind, or Erigon) handles transaction processing and smart contract execution. The consensus client (like Prysm, Lighthouse, or Teku) manages validator attestations and finality.

Full nodes store approximately the last 128 blocks on disk, though this will expand to 8,192 blocks after the May 2025 Pectra upgrade¹⁷. Older data gets pruned automatically to save disk space. While full nodes can regenerate historical states by re-executing transactions, this process is computationally intensive and not practical for frequent historical queries¹⁸.

Archive Nodes: Complete Historical Record Keepers

Archive nodes inherit all capabilities of full nodes while additionally building and maintaining a complete archive of every historical state since the genesis block¹⁹. This comprehensive storage makes archive nodes invaluable for blockchain explorers, analytics platforms, and applications requiring deep historical data access. Unlike full nodes that must recompute old states, archive nodes retrieve historical information instantly from local storage²⁰.

The storage requirements for archive nodes are substantial. As of 2025, a Geth-based archive node requires approximately 15TB of disk space, while more efficient clients like Erigon need around 2TB due to optimized storage architecture²¹. These requirements continue growing as the blockchain expands. Archive nodes are not required to participate in block validation—they serve primarily as historical data providers²².

Light Nodes: Efficient Network Participation

Light nodes provide the most resource-efficient way to interact with Ethereum²³. Instead of storing complete block data, light nodes download only block headers containing summary information like timestamps, previous block hashes, and state roots. When light nodes need additional data, they request it from full nodes and cryptographically verify responses using Merkle proofs²⁴.

This design enables light nodes to run on devices with limited storage and processing power, including personal computers and mobile phones²⁵. Light nodes cannot participate in consensus or serve as validators, but they provide the same security guarantees as full nodes for basic operations like checking balances and verifying transactions²⁶. The cryptographic verification ensures light nodes can detect invalid data even without storing complete blockchain history.

How Ethereum’s Mempool Functions as a Transaction Waiting Room

The Ethereum mempool, short for “memory pool,” serves as a temporary holding area where valid transactions wait before inclusion in blocks²⁷. Understanding mempool mechanics is crucial for miners and users seeking to optimize transaction timing and costs. Each Ethereum node maintains its own mempool rather than sharing a universal pool across the network²⁸.

The mempool is not a single shared database but rather individual waiting areas maintained by each node, with transactions propagating across the network through peer-to-peer gossip protocols.

Transaction Lifecycle Through the Mempool

When a user creates and signs a transaction, it gets broadcast to an Ethereum node²⁹. That node performs validity checks including signature verification, sufficient balance confirmation, and correct nonce sequencing. Valid transactions enter the node’s mempool and propagate to peer nodes, gradually spreading across the network. Validators select transactions from their mempool to include in proposed blocks, typically prioritizing those offering higher gas fees³⁰.

Transactions remain in the mempool until one of several outcomes occurs³¹. Successfully included transactions are removed when their containing block is added to the chain. Transactions can be replaced through “cancel and speed up” operations where users submit new transactions with higher fees and the same nonce. Low-fee transactions may be dropped entirely if the mempool becomes full and nodes need to make space for higher-priority transactions³².

Gas Fees and Transaction Prioritization

Validators prioritize mempool transactions based on effective gas price—the combination of base fee plus priority fee (tip)³³. After EIP-1559’s implementation, transactions specify a maximum fee and priority fee. The base fee gets burned while the priority fee goes directly to validators as an incentive for quick inclusion³⁴.

Research published in 2025 reveals persistent fee-based disparities despite EIP-1559’s intended improvements³⁵. High-fee transactions consistently receive faster processing, while low-fee transactions face delays or exclusion during network congestion. Mempool congestion remains a key factor affecting validator efficiency and proposal latency, with extremely high fees not always guaranteeing faster confirmation due to execution inefficiencies³⁶.

The Mempool-Finality Connection: How They Interact

Mempool conditions directly influence how quickly transactions progress toward finality³⁷. The journey from mempool to finality follows a predictable path: transaction broadcast to mempool, validator selection and block inclusion, block attestation by validators, checkpoint justification after one epoch, and finally checkpoint finalization after two epochs.

Network congestion affects each stage of this process. When the mempool contains high transaction volumes, validators face increased computational load processing attestations and executing transactions³⁸. This additional load can modestly lengthen average block finalization times, particularly during periods of extreme network activity³⁹. Gas price volatility and pending transaction load increase inclusion latency, creating cascading effects through the consensus layer.

Finality Incidents and Mempool Behavior

During finality incidents—periods when the chain temporarily fails to finalize new checkpoints—validators can fall behind due to increased processing demands⁴⁰. This creates a cascade effect where delays compound through the network. Empirical measurements show that in such events, the time between a transaction entering the mempool and its on-chain inclusion increases significantly, degrading user experience even though the mempool continues accepting and relaying transactions normally⁴¹.

The interaction works both ways. Mempool congestion slows finality by increasing validator processing time, while finality delays cause mempool backlogs as fewer blocks get confirmed per unit time. This creates feedback loops during network stress where both systems mutually reinforce degraded performance⁴².

MEV and Mempool Exploitation: What Miners Need to Know

Maximal Extractable Value (MEV) represents profits extractable by manipulating transaction ordering within blocks⁴³. The public mempool’s transparency enables MEV searchers to identify profitable opportunities by observing pending transactions. As of 2025, over $1.8 billion in MEV has been extracted from Ethereum users⁴⁴.

Sandwich Attacks in Detail

Sandwich attacks represent one of the most common MEV exploitation methods⁴⁵. When searchers identify a large pending DEX trade in the mempool, they calculate the expected price impact. The searcher then places a buy order immediately before the target transaction with higher gas fees, causing it to execute first at a lower price. Immediately after the victim’s transaction executes and raises the price, the searcher places a sell order, profiting from the price increase their own sandwich created⁴⁶.

For example, if Alice attempts to buy 10,000 UNI tokens with DAI, a sandwich attacker might buy 5,000 UNI first, let Alice’s transaction execute at the elevated price, then immediately sell the 5,000 UNI at profit. Alice ends up paying significantly more than she would have without the sandwich attack⁴⁷.

Protecting Against MEV Exploitation

Several mitigation strategies have emerged to combat MEV extraction⁴⁸. Private RPCs and mempools like Flashbots Protect route transactions directly to validators without broadcasting to the public mempool, preventing searchers from observing pending transactions⁴⁹. MEV protection services like MEV Blocker redistribute extracted value back to users rather than allowing searchers to capture all profits⁵⁰.

New proposals in 2025 introduce encrypted mempools where transactions remain encrypted until block inclusion, making MEV extraction technically impossible rather than just difficult⁵¹. Shutter Network’s proposal for distributed encrypted mempools on Ethereum aims to implement threshold encryption across multiple keyper nodes, preventing any single party from viewing pending transactions before inclusion⁵².

Hardware Requirements for Running Ethereum Nodes

Operating Ethereum nodes requires substantial hardware investments, with requirements varying significantly by node type⁵³. For miners considering node operation, understanding these requirements helps in infrastructure planning and cost management.

Full Node Hardware Specifications

Running a full Ethereum node in 2025 requires a fast CPU with at least 4 cores, preferably starting at 3.5 GHz or higher⁵⁴. RAM requirements stand at a minimum 16GB, though 32GB is recommended for optimal performance⁵⁵. Storage represents the most demanding requirement—a minimum 2TB SSD is necessary, with 4TB recommended to avoid frequent upgrades as blockchain size grows⁵⁶.

Internet connectivity needs are equally important. A connection providing at least 100 Mbps bandwidth with 30-40TB monthly data transfer capacity ensures nodes can synchronize and propagate blocks efficiently⁵⁷. Traditional hard disk drives (HDDs) are inadequate—the random read/write patterns of blockchain operations require solid-state drives (SSDs) or NVMe storage for acceptable performance⁵⁸.

Archive Node Infrastructure Demands

Archive nodes demand significantly more resources due to complete historical state storage⁵⁹. Storage requirements vary dramatically by client implementation. A Geth-based archive node requires approximately 15TB or more as of 2025, while Erigon’s optimized architecture reduces this to around 2-3TB through efficient data structures and compression⁶⁰. CPU and RAM requirements remain similar to full nodes, though higher specifications improve query performance for historical data access⁶¹.

Light Node Minimal Requirements

Light nodes offer the most accessible entry point for network participation⁶². Storage requirements remain minimal at just a few gigabytes for block headers. Standard consumer hardware including personal computers and even Raspberry Pi devices can run light nodes effectively⁶³. This accessibility makes light nodes ideal for users wanting to verify transactions without investing in dedicated infrastructure.

Practical Applications for Miners and Infrastructure Operators

Understanding Ethereum’s architecture provides concrete benefits for mining operations and infrastructure planning. Node type selection impacts both operational costs and capabilities.

Case Study: Mining Pool Node Architecture

Large mining pools typically operate multiple full nodes across geographic regions for redundancy and optimal network connectivity. F2Pool, one of Ethereum’s major mining pools before The Merge, maintained full nodes in Asia, Europe, and North America to minimize block propagation delays⁶⁴. This distributed architecture ensured their validators could access mempool transactions quickly and propose blocks with minimal latency⁶⁵.

Case Study: Blockchain Analytics Infrastructure

Etherscan, Ethereum’s leading block explorer, relies heavily on archive nodes to serve historical transaction queries⁶⁶. Their infrastructure processes millions of historical data requests daily, requiring extensive archive node deployments. The ability to instantly query any historical state without recomputation makes archive nodes essential for this use case⁶⁷.

Future Developments: Single Slot Finality

Ethereum developers are researching single slot finality (SSF) to reduce the current 15-minute finality time to a single 12-second slot⁶⁸. This improvement would dramatically enhance user experience and enable new applications requiring rapid finality confirmation. However, SSF introduces significant technical challenges around validator computational load and network bandwidth⁶⁹.

Ethereum co-founder Vitalik Buterin outlined his vision for this transformation in his 2025 development priorities: “Single-slot finality represents the holy grail of blockchain performance”, explaining that the proposed upgrade would allow blocks to become final within a single slot rather than requiring two epochs⁷⁰. Buterin emphasized that while software solutions exist to enable SSF, the real challenge lies in implementation without compromising network decentralization⁷¹.

The Pectra upgrade scheduled for May 2025 will implement incremental improvements including raising validator balance limits from 32 ETH to 2,048 ETH, reducing the number of separate validator instances operators must run⁷². This consolidation improves network efficiency while maintaining decentralization through a large, diverse validator set⁷³.

Conclusion

Ethereum’s mempool, finality mechanisms, and node architecture work together to create a secure, decentralized transaction processing system. The mempool serves as the entry point where transactions begin their journey, node operators provide the infrastructure that processes and stores blockchain data, and finality delivers the ultimate guarantee of transaction permanence. For miners and infrastructure operators, mastering these concepts enables optimized gas fee strategies, informed hardware investments, and protection against MEV exploitation. As Ethereum continues evolving toward single slot finality and encrypted mempools, understanding these foundational systems becomes increasingly valuable for anyone operating in the Ethereum ecosystem.

Ethereum Finality Mempool FAQs

How long does Ethereum finality take?

Ethereum finality typically takes approximately 15 minutes or 64-95 slots under normal network conditions⁷². This timeframe allows validators to attest to two epochs of checkpoint blocks, with at least two-thirds of staked ETH voting for finalization. Network congestion can extend this period during high-activity periods.

What is the difference between Ethereum finality and confirmation?

Ethereum confirmation occurs when a transaction is included in a block, providing probabilistic certainty within minutes⁷³. Finality represents absolute certainty—the point when reverting a transaction would require burning at least one-third of all staked ETH, making changes economically impossible⁷⁴. Finality provides stronger guarantees than simple confirmation.

How does the Ethereum mempool differ from Bitcoin’s mempool?

While both serve as transaction waiting areas, Ethereum’s mempool prioritizes transactions using a two-tier gas system with base fees and priority tips introduced by EIP-1559⁷⁵. Bitcoin’s mempool uses simpler fee-per-byte prioritization. Ethereum’s mempool also handles smart contract transactions with varying computational complexity, while Bitcoin primarily processes simple value transfers.

Can you see pending transactions in the Ethereum mempool?

Yes, the Ethereum mempool is generally public and accessible through node RPC interfaces using methods like eth_subscribe⁷⁶. Services like Blocknative and Quicknode provide mempool monitoring APIs that expose pending transaction data in real-time. However, private mempools like Flashbots Protect allow users to submit transactions directly to validators without public mempool exposure⁷⁷.

Why would someone run an Ethereum archive node instead of a full node?

Archive nodes enable instant access to any historical blockchain state since genesis without requiring computational reprocessing⁷⁸. Applications like blockchain explorers, analytics platforms, and services requiring historical data queries benefit from archive nodes despite their higher storage costs. Full nodes must recompute old states from transaction history, making frequent historical queries impractical.

Ethereum Finality Mempool Citations

1. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025. //ethereum.org/developers/docs/consensus-mechanisms/pos/
2. Quicknode. “Ethereum Full Node vs. Archive Node.” Quicknode Guides, 2025. //www.quicknode.com/guides/infrastructure/node-setup/ethereum-full-node-vs-archive-node
3. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025. //ethereum.org/roadmap/single-slot-finality/
4. Ledger Academy. “Ethereum Proof of Stake: Explained.” Ledger, 2025. //www.ledger.com/academy/ethereum-proof-of-stake-pos-explained
5. HackMD. “What Happens After Finality in ETH2?” Prysmatic Labs, 2025. //hackmd.io/@prysmaticlabs/finality
6. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025.
7. Blocknative. “The Ethereum Merge: A Proof-of-Stake Upgrade Guide.” Blocknative Blog, 2025. //www.blocknative.com/blog/ethereum-merge-proof-of-stake
8. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025.
9. Datawallet. “Top 8 Ethereum Staking Statistics and Trends in 2025.” Datawallet, 2025. //www.datawallet.com/crypto/ethereum-staking-statistics-and-trends
10. Ledger Academy. “Ethereum Proof of Stake: Explained.” Ledger, 2025.
11. OneKey. “What Is Ethereum Proof-of-Stake?” OneKey Blog, 2025. //onekey.so/blog/ecosystem/what-is-ethereum-proof-of-stake/
12. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025.
13. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025.
14. Alchemy. “Types of Ethereum Nodes: Full vs. Archive vs. Light.” Alchemy Overviews, 2025. //www.alchemy.com/overviews/full-vs-light-vs-archive-nodes
15. Quicknode. “Ethereum Full Node vs. Archive Node.” Quicknode Guides, 2025.
16. Quicknode. “Ethereum Full Node vs. Archive Node.” Quicknode Guides, 2025.
17. GetBlock. “Light vs Full vs Archive Ethereum Nodes.” GetBlock.io, May 2025. //getblock.io/blog/light-vs-full-vs-archive-ethereum-nodes/
18. Ethereum Foundation. “Nodes and clients.” Ethereum.org, 2025. //ethereum.org/developers/docs/nodes-and-clients/
19. Ethereum Foundation. “Ethereum Archive Node.” Ethereum.org, October 21, 2025. //ethereum.org/developers/docs/nodes-and-clients/archive-nodes/
20. Alchemy. “Archive Nodes – Everything You Need to Know.” Alchemy Overviews, 2025. //www.alchemy.com/overviews/archive-nodes
21. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025. //www.bacloud.com/en/knowledgebase/203/ethereum-node-server-requirements-2025-updated.html
22. Ethereum Foundation. “Ethereum Archive Node.” Ethereum.org, October 21, 2025.
23. Cloudflare. “Node types.” Cloudflare Web3 docs, 2025. //developers.cloudflare.com/web3/ethereum-gateway/concepts/node-types/
24. Alchemy. “Types of Ethereum Nodes: Full vs. Archive vs. Light.” Alchemy Overviews, 2025.
25. CryptoAPIs. “Full, Archive & Light Types Of Ethereum Nodes.” CryptoAPIs, 2025. //cryptoapis.io/blog/127-full-archive-light-types-of-ethereum-nodes
26. Ethereum Foundation. “Nodes and clients.” Ethereum.org, 2025.
27. GeeksforGeeks. “What is Ethereum Mempool?” GeeksforGeeks, July 23, 2025. //www.geeksforgeeks.org/ethical-hacking/what-is-ethereum-mempool/
28. Quicknode. “How to Access Ethereum Mempool.” Quicknode Guides, 2025. //www.quicknode.com/guides/ethereum-development/transactions/how-to-access-ethereum-mempool
29. Blocknative. “What is the Mempool? – How Blockchain Transactions Work.” Blocknative Blog, 2025. //www.blocknative.com/blog/mempool-intro
30. GeeksforGeeks. “What is Ethereum Mempool?” GeeksforGeeks, July 23, 2025.
31. Quicknode. “How to Access Ethereum Mempool.” Quicknode Guides, 2025.
32. Quicknode. “How to Access Ethereum Mempool.” Quicknode Guides, 2025.
33. Ethereum Foundation. “Maximal extractable value (MEV).” Ethereum.org, 2025. //ethereum.org/developers/docs/mev/
34. Hossain, S. M. Mostaq, et al. “Unraveling Ethereum’s Mempool: The Impact of Fee Fairness, Transaction Prioritization, and Consensus Efficiency.” ArXiv, June 9, 2025. //arxiv.org/abs/2506.07988
35. Hossain, S. M. Mostaq, et al. “Unraveling Ethereum’s Mempool.” ArXiv, June 9, 2025.
36. Hossain, S. M. Mostaq, et al. “Unraveling Ethereum’s Mempool.” ArXiv, June 9, 2025.
37. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025.
38. Hossain, S. M. Mostaq, et al. “Unraveling Ethereum’s Mempool.” ArXiv, June 9, 2025.
39. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025.
40. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025.
41. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025.
42. CTOL Digital Solutions. “Vitalik Buterin Unveils Ambitious 2025 Ethereum Roadmap Focused on Single-Slot Finality.” CTOL Digital, 2025. //www.ctol.digital/news/vitalik-buterin-unveils-2025-ethereum-roadmap-single-slot-finality-ecosystem-resilience/
43. CCN. “Vitalik Buterin’s Blueprint for Ethereum Post-Merge: 1 ETH Staking and Instant Finality.” CCN.com, October 28, 2025. //www.ccn.com/news/crypto/vitalik-buterin-blueprint-ethereum-future/
44. Datawallet. “Top 8 Ethereum Staking Statistics and Trends in 2025.” Datawallet, 2025. //www.datawallet.com/crypto/ethereum-staking-statistics-and-trends
45. Datawallet. “Top 8 Ethereum Staking Statistics and Trends in 2025.” Datawallet, 2025.
46. Blocknative. “The Ethereum Merge: A Proof-of-Stake Upgrade Guide.” Blocknative Blog, 2025. //www.blocknative.com/blog/ethereum-merge-proof-of-stake
47. Ledger Academy. “Ethereum Proof of Stake: Explained.” Ledger, 2025.
48. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025.
49. Hossain, S. M. Mostaq, et al. “Unraveling Ethereum’s Mempool.” ArXiv, June 9, 2025.
50. Quicknode. “How to Access Ethereum Mempool.” Quicknode Guides, 2025.
51. Ethereum Foundation. “Maximal extractable value (MEV).” Ethereum.org, 2025.
52. Ethereum Foundation. “Ethereum Archive Node.” Ethereum.org, October 21, 2025.
53. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
54. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
55. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
56. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
57. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
58. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
59. Dysnix. “Archive Nodes: Explanation Based on Use Cases.” Dysnix, 2025. //dysnix.com/blog/archive-nodes
60. Bacloud. “Ethereum node server requirements 2025 updated.” Bacloud, 2025.
61. Dysnix. “Archive Nodes: Explanation Based on Use Cases.” Dysnix, 2025.
62. Alchemy. “Types of Ethereum Nodes: Full vs. Archive vs. Light.” Alchemy Overviews, 2025.
63. CryptoAPIs. “Full, Archive & Light Types Of Ethereum Nodes.” CryptoAPIs, 2025.
64. Blocknative. “The Ethereum Merge: A Proof-of-Stake Upgrade Guide.” Blocknative Blog, 2025.
65. Blocknative. “The Ethereum Merge: A Proof-of-Stake Upgrade Guide.” Blocknative Blog, 2025.
66. Alchemy. “Archive Nodes – Everything You Need to Know.” Alchemy Overviews, 2025.
67. Ethereum Foundation. “Ethereum Archive Node.” Ethereum.org, October 21, 2025.
68. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025.
69. Ethereum Foundation. “Single slot finality.” Ethereum.org, October 21, 2025.
70. Datawallet. “Top 8 Ethereum Staking Statistics and Trends in 2025.” Datawallet, 2025.
71. Datawallet. “Top 8 Ethereum Staking Statistics and Trends in 2025.” Datawallet, 2025.
72. Blocknative. “The Ethereum Merge: A Proof-of-Stake Upgrade Guide.” Blocknative Blog, 2025.
73. Ledger Academy. “Ethereum Proof of Stake: Explained.” Ledger, 2025.
74. Ethereum Foundation. “Proof-of-stake (PoS).” Ethereum.org, 2025.
75. Hossain, S. M. Mostaq, et al. “Unraveling Ethereum’s Mempool.” ArXiv, June 9, 2025.
76. Quicknode. “How to Access Ethereum Mempool.” Quicknode Guides, 2025.
77. Ethereum Foundation. “Maximal extractable value (MEV).” Ethereum.org, 2025.
78. Ethereum Foundation. “Ethereum Archive Node.” Ethereum.org, October 21, 2025.