What is a memory pool and how does it work?

Latest updates on memory pool technology in 2025

Real-time memory pool data analysis

In 2025, memory pool monitoring has significantly developed, with the platform capturing over 7 billion Ethereum memory pool events each month—continuously about 2,000 events per second. This development, referred to by experts as the “dark forest,” has made previously exclusive memory pool data accessible to the public, enabling ordinary traders to monitor operations that were once only visible to advanced players.

• Evolution of Transaction Verification The node now performs comprehensive validation checks, including signature verification, input-output balance confirmation, and prevention of double spending.
• dynamic characteristics The memory pool changes at sub-second intervals, creating a constantly evolving pre-consensus environment.

memory pool architecture improvement

Each node maintains its own unique memory pool configuration, rather than sharing a common pool:

• Resource allocation Low-end devices allocate the least memory for transaction logs, while high-performance nodes invest a lot of resources.
• Transaction Management When nodes receive validated blocks, they will automatically remove the included transactions and transactions with conflicting inputs from their memory pool.

2025 Memory Pool Key Indicators

Transaction Impact

Advanced memory pool monitoring has changed the trading strategies across various platforms. Memory pool browser technology now allows users to detect hostile actions before they reach the blockchain, providing critical advance warning. Similar to traditional finance, cryptocurrency trading increasingly revolves around latency advantages, and memory pool visibility offers a significant competitive edge.

With the development of gateway protocols, the transaction priority mechanism continues to evolve alongside the capabilities of the memory pool, solidifying the fundamental role of the memory pool in blockchain infrastructure, while providing users with unprecedented visibility into transaction processing dynamics.

What is a memory pool, and how do they work?

[TL; DR]

The main function of the memory pool is to store transactions before they are validated.

Bitcoin and Ethereum is an example of a blockchain that uses a memory pool.

Blockchain congestion is the main reason for transactions being stuck for a long time.

If you want the transaction to proceed quickly, please use the appropriate gas fees.

Introduction

We often conduct transactions on various blockchains, such as Ethereum
And Bitcoin, but we may not fully understand some of their technical aspects. However, understanding how certain components of blockchain work may help speed up our transactions or reduce their costs.

In this analysis, we discuss the memory pool in blockchain and its working principles. We will also explain how to handle pending or delayed transactions on the Bitcoin and Ethereum networks.

Reading also: What is a blockchain explorer?

What is the memory pool?

memory pool The abbreviation of memory pool
It is a node space used to store unconfirmed transactions and other information. It serves as a waiting room for transactions that have not yet been confirmed or approved. Remember, only after a transaction has been verified or approved will it be included in a block.

In other words, the memory pool is a space where pending transactions wait before being added to the block.

When a transaction is verified, it means that all nodes on the blockchain can access the information. However, before the transaction is approved, each node will verify the signature, check that the outputs do not exceed the inputs, and ensure that the system has not already sent the digital assets. If any of these conditions are not met, the validator or miner will reject the transaction.

It is important to understand that there is no single memory pool for the blockchain. Therefore, each node on the blockchain network has its own memory pool, also known as the transaction pool. Thus, the memory pools of individual nodes constitute the collective blockchain memory pool.

For example, transactions that occur on the Gate platform. Bitcoin
Transactions on the network are not directly added to the blockchain. Each node stores the transactions in a temporary transaction storage area (referred to as the memory pool or txpool). Therefore, they form a queue that needs to be checked and validated.

Therefore, we can simply define the memory pool as an ordered queue of transactions that need to be sorted and added to a block.

Memory Pool in Nodes - Babypips

Please remember that transactions only exist in the memory pool before they are confirmed or packed into a block. In the chart above, the blue section represents the memory pool space that exists within the node. The next chart shows the distribution of the memory pool in the network.

Full nodes in the Bitcoin network have a memory pool - Babypips

As shown in the above figure, each full node in networks such as Bitcoin or Ethereum has a memory pool. However, each transaction will be broadcasted among the network nodes, as illustrated in the figure below.

Bitcoin Node Network - Babypips

As we can see in the diagram, the Bitcoin transactions initiated by the purple users will be stored in the memory pool before review. It is worth noting that once the transactions are verified, they will be added to a block as shown in the next illustration.

Add transaction to file (block) - Babypips

The blue arrow in the image shows what happens when a transaction is validated; it moves from the memory pool into a block. At this stage, all computers will have an updated file of the confirmed transactions.

How do memory pools work?

Typically, the way memory pools operate is not entirely the same. Their operation depends on the architecture of each blockchain. In other words, how they work depends on the type of consensus mechanism of the network. Transactions are validated either by validating nodes or by miners, depending on the type of network.

For example, transactions on the Bitcoin blockchain are validated by miners because it uses a proof-of-work consensus mechanism. On the other hand, transactions on the Ethereum network are approved by validators because the blockchain uses a proof-of-stake consensus mechanism.

Read more: Reshaping Ethereum and Potential Network Health Issues

Each transaction on the Bitcoin network requires at least six confirmations, while the Ethereum blockchain requires at least seven confirmations. Essentially, every transaction in the memory pool has a specific status. It is either pending or queued. When a validator or miner verifies a queued transaction, it becomes a pending transaction.

Read also: What is the Bitcoin Lightning Network?

Let’s look at a simple example of what happens when you send ETH to a friend. The process involves entering the wallet address, setting the gas fee, and
Confirm Transaction.

Once you perform these actions, your transaction will enter the queue, waiting for verification. At that point, it will be broadcast to the entire node network, but it has not yet become part of any block.

If a validator checks a transaction and finds that it meets all the conditions, it will move from the queued state to the pending state and wait for confirmation. Then, one of the validators will extract the transaction from the memory pool and add it to the new block.

At that stage, several validators approved this, allowing your friend to receive ETH.

Let us summarize the dynamics of the memory pool.

Step 1: The sender enters the target address, accepts the network transaction fee, and clicks “Send”.

STEP 2: The transaction is added to the latest memory pool and queued.

Step 3: The memory pool data is broadcasted to other nodes in the network.

Step 4: A node has approved the transaction, changing its status from queued to pending.

Step 5: Miners select pending transactions and add them to a block.

Step 6: The remaining nodes remove the transaction from their memory pool.

Step 7: Transaction completed, assets enter the target wallet.

Now that you understand the reasons for delays in processing transactions, let’s take a look at some reasons why transactions may get stuck and what actions you can take.

The reason for transactions being stuck in the memory pool

The main reason transactions get stuck in the memory pool is network congestion. When the transaction volume is high, the available block space is under pressure, leading to a backlog in the memory pool. This occurs when the number of pending transactions exceeds the average number of transactions in a block.

Some events and news that affect the level of blockchain traffic may lead to network congestion. For example, airdrops, listings on more exchanges, or new partnerships may cause congestion. However, there are some factors that can cause nodes to prioritize certain transactions.

transaction fees
The fees attached to different transactions determine which transactions get approved first. Since miners and validators are driven by profit, they prioritize transactions with higher fees as this allows them to earn greater rewards.

Therefore, miners and validators organize transactions in their memory pool based on the fee per byte (e.g., satoshis/byte). In the end, they add the transactions with the highest fees to the block first. This system forces users to set high fees for their transactions.

Nevertheless, depending on network traffic, users can set the optimal gas fees and still be able to confirm their transactions in a timely manner. It is important for users to check historical data, network congestion, transaction volume, and fee distribution in the memory pool before setting their own gas fees.

Confirmation Time
Before setting the fees, users may also need to check the confirmation time. When the confirmation time is longer, it means that miners and validators prioritize transactions with high fees. Essentially, when the memory pool is congested, transactions with low fees will take longer to complete.

block space
The block space also determines the time it takes for a transaction to get approved. Since block space is always limited, miners and validators prioritize transactions with higher fees. This means that transactions with lower fees spend more time in the memory pool.

memory pool size and eviction
Each transaction to be added to the block is a data block measured in kilobytes. Therefore, a large memory pool can accommodate many transactions. Typically, nodes set the size of their memory pool, which is usually 300MB for Bitcoin. When the memory pool reaches a threshold, nodes may set a minimum transaction fee.

When users understand this dynamic, they can easily set transaction fees and get approval in a reasonable time. For example, they can avoid overpaying during low throughput periods, or avoid underpaying when making time-sensitive transactions during peak hours. They can also almost accurately guess the speed at which their transactions will be validated.

Why do transactions get stuck in the memory pool?

In many cases, transactions get stuck in the memory pool for various reasons, but most are related to gas fees. However, the biggest reason some transactions get stuck in the memory pool is network congestion. When congestion occurs, transactions with higher fees are processed first, while the rest get stuck there.

Therefore, increasing transaction fees during network peak times is the best choice. This is because gas fees will surge during congestion. Therefore, if you are processing transactions during busy periods, you should match your fees with what other network users are paying.

Another reason transactions get stuck in the memory pool is a drop in hash rate. When the network does not have enough physical computing power to process a large number of pending transactions, we refer to it as a drop in hash rate.

How to release your transaction from the memory pool

When your transaction is stuck in the memory pool, you have several options. The first option is to cancel the transaction. In this case, you will need to pay another transaction fee. For example, you can resend the transaction using the same nonce.

Secondly, you can choose to speed up the transaction by connecting your Ledger to a third-party wallet like MetaMask or Electrum. However, you will need to pay higher fees.

The third option is to wait a little longer for the transaction to complete smoothly. Doing this allows you to wait for the network traffic to decrease, enabling your transaction to be completed without additional fees.

memory pool in Bitcoin

Transactions sent to the Bitcoin network are not immediately added to the blockchain. They first enter the memory pool. In the past, such transactions were measured only by the fee in satoshis per byte.

However, after the SegWit upgrade, this situation changed, allowing transactions to be measured in weight units. The benefit is that the SegWit feature allows more data to be added to a block than before. Specifically, each block has 2MB of data, although it can increase to 4MB.

Ethereum memory pool

Initially, the Ethereum network needed to add transactions to the memory pool before verification. However, when the blockchain transitioned from the proof-of-work consensus mechanism to the proof-of-stake mechanism, the concept of block producers was introduced.

Basically, block builders are third-party participants who compile transactions into optimized bundles to form a block. In this case, the third-party entity compiles or reorders certain transactions from the memory pool into bundles. They then provide the transaction bundles to validators to be included in the block. Third-party entities are rewarded for completing these tasks.

Conclusion

In summary, the memory pool is the space where blockchain nodes temporarily store transactions before they are validated. In Ethereum, the time a transaction stays in the memory pool depends on various factors such as network congestion and gas fees. Both Bitcoin and Ethereum blockchains use a transaction pool.

Frequently Asked Questions about the memory pool

What is the memory pool?

The memory pool refers to the space where unprocessed transactions wait before being added to a block. Once a transaction is validated, it will be added to a block, and this block exists on the blockchain.

What happens in the memory pool?

When some transactions are in the memory pool, some validators or miners will choose them and add them to the block. However, if the network is congested, some transactions may wait in the queue or be suspended for a long time.

What is the BTC memory pool?

The BTC memory pool is the waiting area for transactions processed on the Bitcoin network. Typically, each Bitcoin blockchain node has its own memory pool.

What is the memory pool in Ethereum?

The memory pool in Ethereum is the space for storing transactions that are held on Ethereum nodes before confirmation. The time before these transactions are confirmed depends on the traffic of the Ethereum network during a certain period.