Filecoin: Node Types, Mempool & Finality Explained

Key Takeaways

• Filecoin operates four primary node types based on services provided: Chain Verifier nodes validate blocks, Client nodes interact with storage markets, Storage Miner nodes produce blocks and earn rewards, and Retrieval Miner nodes serve stored data.
• The F3 (Fast Finality) upgrade activated in April 2025 reduced transaction finality from 7.5 hours to minutes (typically 2-10 epochs), delivering a 450-fold improvement that enables DeFi, smart contracts, and cross-chain bridges on Filecoin.
• Filecoin’s mempool is decentralized with each node maintaining its own local message pool that propagates via GossipSub, where storage providers select messages based on gas fees to maximize block rewards.
• Tipsets instead of individual blocks allow multiple storage providers to produce valid blocks per 30-second epoch, increasing network throughput while maintaining decentralization.
• The GossiPBFT consensus protocol works alongside Expected Consensus to provide deterministic finality, making Filecoin viable for latency-sensitive applications previously impossible with hour-long confirmation times.

Article Summary

Filecoin uses specialized node types (Chain Verifier, Client, Storage Miner, and Retrieval Miner) that each serve distinct network functions, while the April 2025 F3 upgrade dramatically shortened transaction finality from 7.5 hours to minutes (typically 2-10 epochs) using GossiPBFT consensus. The network’s decentralized mempool system lets each node maintain local message pools that propagate via GossipSub, with storage providers selecting high-fee messages for inclusion in 30-second epoch blocks organized into tipsets.

Understanding Filecoin Node Types and Their Roles

Unlike Bitcoin’s simple “full node” classification, Filecoin defines nodes based on the specific services they provide to the network. This modular approach lets operators choose exactly which functions their infrastructure supports, from basic blockchain validation to full storage provider operations. Each node type builds upon foundational services while adding specialized capabilities for different network participants.

Chain Verifier Nodes: The Foundation Layer

Chain Verifier nodes represent the minimum requirement for participating in the Filecoin network. These nodes run the Blockchain system, which means they synchronize with the network, validate incoming blocks and tipsets, and maintain consensus about the current chain state. However, Chain Verifier nodes cannot mine blocks or produce their own contributions to the chain. They simply verify that other participants are following protocol rules correctly. Any node participating in Filecoin must provide chain verification services as a baseline.

Think of Chain Verifier nodes as read-only participants that watch the network and keep everyone honest. They download the blockchain, check that each new tipset follows consensus rules, and reject invalid blocks from malicious or misconfigured storage providers. For developers building applications on Filecoin, running a Chain Verifier node provides trustless access to network state without the computational overhead of mining operations.

Client Nodes: Market Participation

Client nodes extend Chain Verifier functionality by adding storage and retrieval market capabilities. These nodes implement both the Storage Market Client and Retrieval Market Client services, letting them negotiate deals with storage providers, upload data for storage, and retrieve stored content when needed. Client nodes also handle data transfers through Filecoin’s Data Transfer Module, managing the actual movement of files across the network.

Exchanges, decentralized applications, and services that build on Filecoin typically operate Client nodes. For example, a decentralized storage application would run a Client node to broker storage deals on behalf of its users, handle file uploads, and coordinate retrievals. The Client node maintains connections to both storage and retrieval markets, tracking available providers and negotiating terms for data storage and access. With the new F3 fast finality system, Client nodes can now confirm storage deals in minutes rather than hours, dramatically improving user experience for applications requiring quick confirmation times.

Storage Miner Nodes: Block Production and Data Storage

Storage Miner nodes represent full network participants that validate blocks, produce new blocks, store client data in sealed sectors, and earn block rewards. These nodes implement the complete Blockchain system plus Storage Mining and Storage Market Provider subsystems. To participate as a Storage Miner, operators must commit at least 10 TiB of storage space and continuously prove they’re storing data correctly through WindowPoSt (Window Proof-of-Spacetime) and WinningPoSt challenges.

Every 30-second epoch, Filecoin’s Expected Consensus mechanism elects storage providers proportional to their storage power. Elected miners can propose new blocks, collect transaction fees, and earn block rewards. Storage Miners pull messages from their local mempool, selecting transactions that maximize their fee earnings while respecting block size limits. The more storage power a miner controls, the higher their probability of election in each epoch. This incentivizes miners to store as much verified client data as possible, growing the network’s useful storage capacity.

Retrieval Miner Nodes: Specialized Data Serving

Retrieval Miner nodes focus exclusively on serving stored data to clients requesting retrievals. Unlike Storage Miners that handle long-term data preservation and block production, Retrieval Miners participate in the retrieval market as content providers. They implement the Retrieval Market Provider service and Data Transfer Module but don’t need to mine blocks or commit storage sectors for consensus participation.

This specialization makes sense for operators who want to provide fast data access without the capital requirements of storage mining. Retrieval Miners can cache popular content, maintain high-speed connections, and optimize their infrastructure for data delivery rather than proof computation. They earn fees by serving retrieval requests, creating a separate economic incentive structure focused on data availability rather than storage commitment.

How Filecoin’s Mempool System Works

The mempool—short for “message pool” or “mpool” in Filecoin terminology—handles all pending transactions waiting for inclusion in the blockchain. Unlike centralized systems with a single queue, Filecoin’s mempool is fundamentally decentralized. Each node maintains its own local message pool that represents messages it wants to see included on-chain. There’s no global central mempool that all participants share, which means different nodes may temporarily see different sets of pending transactions.

Message Propagation via GossipSub

When a user or application broadcasts a new message to the Filecoin network, that message enters the local mempool of the receiving node. The node then propagates the message to its peers using libp2p’s GossipSub protocol, which efficiently distributes messages across the peer-to-peer network without flooding every connection. GossipSub uses probabilistic gossiping where nodes select random subsets of peers to forward messages, ensuring rapid network-wide distribution while maintaining manageable bandwidth requirements.

As messages propagate through the network, each receiving node independently validates them before adding to its local mempool. Invalid messages—those with incorrect signatures, insufficient gas, or other protocol violations—get rejected without propagation. This distributed validation prevents spam and malformed transactions from consuming network resources. Within seconds, a valid message typically reaches most storage providers’ mempools, making it available for inclusion in upcoming blocks.

Gas Fees and Message Selection

Storage providers face an optimization problem when selecting which messages to include in their blocks. Each message specifies a GasFeeCap (maximum fee per gas unit) and GasPremium (priority fee paid directly to the block producer). Messages also declare a GasLimit indicating expected computation cost. Storage providers run sophisticated selection algorithms that maximize their earnings while respecting the block’s gas capacity limits.

The selection process works like an auction where messages compete via their gas pricing. Higher GasPremium values increase a message’s attractiveness to storage providers because that portion goes directly to the miner. Part of each message’s total gas cost gets burned by the protocol, creating deflationary pressure on FIL supply. Messages with insufficient gas pricing may sit in mempools for extended periods until network congestion decreases or users increase their fee caps.

If your transaction gets stuck in Filecoin’s mempool, you can replace it by resubmitting with the same nonce but higher gas fees. This mechanism lets users “unstick” low-priority messages by increasing their economic incentive for inclusion. However, the replacement must offer at least 1.25 times the original gas premium to be accepted into the mempool—a rule that prevents spam from free message replacement attacks.

Mempool Management and Cleaning

Nodes periodically clean their local mempools to prevent unbounded memory growth. The cleaning algorithm prioritizes removing messages with the lowest gas fees first, since these have minimal chance of inclusion anyway. There’s no fixed time limit for how long messages can persist in the mempool, but practical considerations mean very low-fee messages eventually get purged during routine maintenance.

Operators can configure mempool parameters like size limits, fee ratios for replacement, and pruning intervals through their node software. The Lotus implementation provides RPC methods for inspecting mempool contents, manually pushing messages, and adjusting configuration parameters. Storage providers often tune these settings based on their specific hardware, network conditions, and strategy for message selection.

Filecoin Finality: From Expected Consensus to F3

Expected Consensus: The Original Mechanism

Filecoin’s original consensus system, Expected Consensus (EC), operates as a heaviest-chain protocol similar to Bitcoin’s longest-chain rule. Every 30 seconds marks a new epoch where the network probabilistically elects multiple storage providers to propose blocks based on their storage power. These elected miners can each create valid blocks during their epoch, and all valid blocks from the same epoch with the same parents form a “tipset”—Filecoin’s fundamental unit of chain growth.

The tipset architecture provides several advantages over single-block systems. Multiple storage providers can contribute work simultaneously without creating conflicts, increasing network throughput and efficiency. All valid work gets rewarded rather than wasting computational effort on orphaned blocks. The system incentivizes collaboration and prevents centralization by rewarding more participants per epoch. Because tipsets require matching parents and heights, the chain achieves rapid convergence during network forks.

However, Expected Consensus only provides probabilistic finality. A tipset becomes more secure as subsequent tipsets build upon it, but no definitive threshold marks guaranteed immutability. For practical purposes, Filecoin clients considered blocks final after 900 epochs—approximately 7.5 hours. This conservative threshold ensured safety against chain reorganizations but created frustrating delays for applications requiring quick confirmation.

The F3 Revolution: GossiPBFT Fast Finality

Fast Finality in Filecoin (F3) represents one of the most transformative upgrades since mainnet launch. Activated on April 29, 2025 at epoch 4920480, F3 introduces a secondary consensus mechanism called GossiPBFT (Gossip-Based Byzantine Fault Tolerant Protocol) that works alongside Expected Consensus. While EC continues handling block production and chain growth, GossiPBFT adds a finality layer that deterministically confirms tipsets within minutes.

GossiPBFT operates through a voting process where storage providers weighted by their power participate in finalizing tipsets. When more than two-thirds of the network’s storage power votes to finalize a specific tipset, that tipset achieves deterministic finality and cannot be reorganized. The protocol uses lightweight cryptographic messages that propagate quickly through the network, typically finalizing tipsets within 2-3 communication rounds. Under normal network conditions, F3 achieves finalization in the same epoch the tipset is proposed.

The protocol design ensures Byzantine fault tolerance, meaning it remains secure even if up to one-third of storage power behaves maliciously or experiences failures. GossiPBFT provides agreement (all honest participants finalize the same tipset), validity (finalized tipsets are valid under EC rules), and termination (the protocol completes within finite time). Unlike some finality gadgets that can stall indefinitely under adversarial conditions, GossiPBFT includes recovery mechanisms that maintain liveness.

Practical Impact of Fast Finality

The reduction from 7.5 hours to minutes fundamentally changes what applications can build on Filecoin. Cross-chain bridges previously faced impractical waiting periods that made asset transfers tedious and capital-inefficient. Now bridges can finalize transactions in minutes, enabling smooth interoperability between Filecoin and other blockchain ecosystems. Decentralized finance applications benefit from quick settlement times that make lending, trading, and yield farming viable on Filecoin’s infrastructure.

Smart contracts running on Filecoin Virtual Machine (FVM) can now offer responsive user experiences comparable to established DeFi chains. Exchanges can reduce deposit confirmation requirements from hundreds of epochs to just a few minutes, improving user satisfaction and capital efficiency. Storage deal confirmation speeds up dramatically, letting clients see cryptographic proof of storage in near real-time rather than waiting hours for conservative finality thresholds.

According to Messari’s Q2 2025 report, total data stored on Filecoin through active deals reached 1,100 PiB in Q2 2025. This represents an 11% decline from Q1 2025’s 1,300 PiB, reflecting the network’s strategic shift from maximizing raw storage volume to emphasizing high-value, enterprise-oriented storage solutions. By the end of Q2, Filecoin had onboarded 2,416 datasets, including 864 large datasets exceeding 1,000 TiB. Industry experts attribute this quality-focused growth partially to F3’s activation, which made Filecoin viable for latency-sensitive enterprise applications. The Filecoin Foundation noted in their April 2025 quarterly update that AI projects including Recall, Theoriq, and Ungate AI are leveraging Filecoin storage specifically because F3 provides the transaction speeds required for production deployment.

How Mempool and Finality Work Together

Understanding how mempool mechanics and finality interact clarifies Filecoin’s transaction lifecycle. When a user submits a message, it first propagates through mempools across the network via GossipSub. Storage providers racing to produce the next block pull messages from their local mempools, selecting high-value transactions that maximize fee earnings. Once a storage provider includes a message in their block and that block joins a tipset, the message is “on-chain” but not yet final.

Under the legacy Expected Consensus system, applications would wait many epochs before treating transactions as irreversible. Conservative applications might wait the full 900 epochs to guarantee safety against chain reorganizations. More aggressive applications might accept 30-60 epoch confirmations, trading some security for faster user experience. This variability created inconsistent behavior across the ecosystem and confused users about true transaction status.

With F3 active, the finality process follows a clearer path. A message enters the mempool, gets included in a block within one or two epochs depending on gas pricing, and achieves deterministic finality within minutes through GossiPBFT voting. Applications can treat F3-finalized tipsets as absolutely immutable, eliminating guesswork about confirmation thresholds. The dual consensus system provides both EC’s throughput benefits and GossiPBFT’s strong finality guarantees.

During network congestion or connectivity issues, F3 finalization might lag behind chain progression. The Filecoin chain continues growing through Expected Consensus while GossiPBFT catches up with voting rounds. This design ensures liveness—the network keeps processing transactions even if finality temporarily slows. Once connectivity improves or participation increases, F3 rapidly finalizes the backlog of pending tipsets.

Node Types Comparison Table

Real-World Case Study: Cross-Chain Bridges After F3

Before F3 activation, cross-chain bridges connecting Filecoin to networks like Ethereum faced a critical challenge. Bridge operators needed absolute certainty that Filecoin transactions were irreversible before releasing wrapped assets on destination chains. Waiting 7.5 hours for conservative finality created terrible user experience—imagine depositing FIL to receive wFIL on Ethereum and waiting most of a work day for confirmation.

The economic risks were equally severe. Bridge operators had to lock significant capital to cover the time delay between receiving Filecoin transactions and final settlement. If users demanded instant bridging, operators absorbed the risk of potential chain reorganizations during the 900-epoch window. This capital inefficiency made bridges expensive to operate and forced users to pay higher fees to compensate for operator risk.

After F3’s April 2025 activation, bridge behavior transformed dramatically. Operators can now achieve deterministic finality within minutes using GossiPBFT confirmation. Users deposit FIL, see it finalized through F3 within minutes, and receive their wrapped tokens on the destination chain almost immediately. The capital efficiency improvements let bridge operators reduce fees, while the improved user experience drives adoption. In Q4 2024, Filecoin saw total lending market inflows of $99 million, marking a 21% increase QoQ from $82 million in Q3 2024, suggesting renewed demand for FIL-backed loans and liquidity amid market volatility. The F3 upgrade’s reduction in finality times is expected to significantly improve cross-chain bridge efficiency by enabling faster transaction settlement and reducing capital lockup periods for bridge operators.

Technical Implementation Notes for Crypto Miners

If you’re running Filecoin infrastructure as a crypto miner or storage provider, understanding these systems helps optimize your operations. When F3 activated on mainnet, node operators needed to upgrade to Lotus v1.32.3 or manually set the F3_INITIAL_POWERTABLE_CID environment variable to participate in fast finality. Failing to upgrade meant your node couldn’t participate in GossiPBFT voting rounds, reducing your effective influence on finalization.

For mempool management, consider monitoring your local message pool size and adjusting pruning parameters based on network conditions. During high congestion periods, increasing your mempool size limits helps capture more high-value messages for inclusion in your blocks. However, larger mempools consume more memory and increase message validation overhead. Finding the right balance depends on your hardware capabilities and profit optimization strategy.

Storage miners should implement automated gas estimation strategies that analyze mempool conditions and adjust fee recommendations dynamically. Tools like the Lotus CLI provide commands for inspecting pending messages, estimating gas costs, and manually adjusting message parameters. The MpoolPushMessage RPC method handles nonce assignment, message signing, and mempool submission atomically, reducing the risk of nonce conflicts when submitting multiple messages.

F3 bandwidth considerations matter for operators with limited connectivity. During catch-up periods when F3 finalizes backlogged tipsets, message sizes increase because each voting message contains information about the entire chain segment being finalized. Monitor the f3/gpbft and f3/certexchange log levels to track bandwidth usage patterns. If you notice sustained high bandwidth linked to F3 operations, consider upgrading your network infrastructure or adjusting participation settings to manage resource consumption.

Conclusion

Filecoin’s architecture combines specialized node types, a decentralized mempool system, and dual consensus mechanisms to deliver a robust decentralized storage network. The April 2025 F3 upgrade represents a watershed moment, transforming Filecoin from a pure storage network into a platform capable of hosting DeFi, cross-chain bridges, and latency-sensitive applications. By understanding how Chain Verifier, Client, Storage Miner, and Retrieval Miner nodes interact, how messages propagate through decentralized mempools, and how GossiPBFT delivers deterministic finality in minutes rather than hours, you can make informed decisions about participating in Filecoin’s ecosystem. Whether you’re running storage infrastructure, building applications, or simply storing data, these technical foundations directly impact your success on the network. The combination of improved finality times and existing storage capabilities positions Filecoin uniquely in the blockchain landscape—fast enough for modern applications, secure enough for valuable assets, and purpose-built for the data economy.

Filecoin Finality Mempool FAQs

How does Filecoin finality mempool work with F3?

Filecoin finality mempool operates through a dual system where messages enter local mempools via GossipSub propagation, get selected by storage providers based on gas fees, and achieve deterministic finality through F3’s GossiPBFT protocol within minutes (typically 2-10 epochs). The mempool handles pending transactions while F3 confirms included messages are irreversible, creating a seamless flow from submission to final confirmation.

What’s the difference between Filecoin node types?

Filecoin node types differ by services provided: Chain Verifier nodes only validate blocks, Client nodes add market participation for storage deals, Storage Miner nodes produce blocks and earn rewards, and Retrieval Miner nodes serve stored data. Each type builds on the previous with additional capabilities, letting operators choose their level of network participation based on resources and goals.

How long does Filecoin finality take now?

Filecoin finality now takes minutes (typically 2-10 epochs) with F3 activated, compared to the legacy 7.5 hours required under Expected Consensus alone. This 450-fold improvement makes Filecoin viable for DeFi, smart contracts, and cross-chain bridges that require fast transaction confirmation.

Can messages get stuck in Filecoin’s mempool?

Yes, messages can get stuck in Filecoin’s mempool if gas fees are too low during network congestion. Users can unstuck messages by resubmitting with the same nonce but at least 1.25 times higher gas premium, which replaces the original message and incentivizes storage providers to include it in the next block.

What is GossiPBFT in Filecoin?

GossiPBFT (Gossip-Based Byzantine Fault Tolerant Protocol) is Filecoin’s F3 finality mechanism that works alongside Expected Consensus to provide deterministic transaction confirmation. Storage providers weighted by power vote on tipsets, and when two-thirds agree, that tipset achieves irreversible finality within minutes, enabling fast-settling applications on Filecoin.

Filecoin Finality Mempool Citations

1. Filecoin Documentation – Node Providers: https://docs.filecoin.io/nodes/full-nodes/node-providers
2. Filecoin Specification – Node Types: https://spec.filecoin.io/systems/filecoin_nodes/node_types/
3. Lotus Documentation – Node Types: https://lotus.filecoin.io/lotus/get-started/use-cases/
4. Filecoin Foundation – How F3 is Transforming the Filecoin Network: https://filecoin.io/blog/posts/how-f3-is-transforming-the-filecoin-network/
5. GitHub – go-f3 Implementation: https://github.com/filecoin-project/go-f3
6. Crypto News Flash – Filecoin Brings Fast Finality to Mainnet with F3 Upgrade: https://www.crypto-news-flash.com/filecoin-brings-fast-finality-to-mainnet-with-f3-upgrade/
7. Gate.com – What is Fast Finality (F3)?: https://www.gate.com/learn/articles/what-is-fast-finality-f3/5666
8. Filecoin – NV25 Teep Network Upgrade: https://filecoin.io/blog/posts/filecoins-nv25-teep-network-upgrade-is-here/
9. Filecoin Specification – Message Pool: https://spec.filecoin.io/systems/filecoin_blockchain/message_pool/
10. Lotus Documentation – Message Pool Management: https://lotus.filecoin.io/lotus/manage/message-pool/
11. Filecoin Documentation – Blocks and Tipsets: https://docs.filecoin.io/basics/the-blockchain/blocks-and-tipsets
12. Filecoin Specification – Storage Mining Cycle: https://spec.filecoin.io/systems/filecoin_mining/storage_mining/mining_cycle/
13. Filecoin FIP-0086 – Fast Finality Specification: https://github.com/filecoin-project/FIPs/blob/master/FIPS/fip-0086.md
14. Medium – Finality Unveiled by FilOz: https://medium.com/@filoz/finality-unveiled-passive-testing-to-mainnet-launch-of-f3-fast-finality-03e09bc68de5
15. Filecoin Foundation – Fresh From FF May 2025: https://fil.org/blog/fresh-from-ff-may-2025
16. Messari – State of Filecoin Q2 2025: https://messari.io/report/state-of-filecoin-q2-2025
17. Messari – State of Filecoin Q4 2024: https://messari.io/report/state-of-filecoin-q4-2024
18. Filecoin Foundation Quarterly Update: April 2025: https://fil.org/blog/filecoin-foundation-quarterly-update-april-2025
19. Filecoin Documentation – Exchange Integration: https://docs.filecoin.io/reference/exchanges/exchange-integration
20. Filecoin Specification – Tipset Structure: https://spec.filecoin.io/systems/filecoin_blockchain/struct/tipset/