Blockchain infrastructureالبنية التحتية لسلاسل الكتل

Chains are the networks transactions get recorded on, and bridges are what carry value from one chain to another. They are not all alike: they differ in cost, in speed, and in who guarantees a transaction is final.

They are not details for developers alone. The choice of network decides what a single transaction costs, how long you wait, and what happens when something breaks. Those are product decisions, not engineering ones.

Who gets to write in the ledger

Start with a simple question: if no bank owns the ledger, who decides which transaction gets recorded?

The answer is the consensus mechanism (آلية الإجماع). The first form of it is proof of work (إثبات العمل), where machines compete to solve an expensive computation and whoever solves it earns the right to write the next block. Its security comes from its cost: forging the ledger would require more computing power than the entire network. Bitcoin still runs this way, and the price is enormous electricity consumption.

The second form is proof of stake (إثبات الحصة), where participants deposit a balance as a bond and whoever cheats loses it. Ethereum moved to this in September 2022 in what was called the Merge, and its energy use fell by more than ninety-nine percent.

Layer 1 and its limits

The base network where final settlement happens is a Layer 1 (الطبقة الأولى). Bitcoin, Ethereum and Solana are examples, and they differ sharply in the ways a product cares about.

Bitcoin is deliberately slow, designed for holding value rather than small payments. Ethereum is slower and more expensive than Solana, but it is older, more scrutinised, and has deeper liquidity. Solana is much faster and cheaper, and has suffered full network outages in past years, a failure mode a financial institution does not accept easily.

The price of a single transaction is the gas fee (رسوم الغاز). The important thing about it is that it is not fixed. It rises with congestion. So a product built on small transactions can suddenly become impossible to price on a busy day, which is a real operational risk rather than a theoretical one.

Layer 2

To solve the cost problem, Layer 2 (الطبقة الثانية) appeared, most commonly as rollups (التجميعات) such as Arbitrum, Base and Optimism.

The idea is to execute thousands of transactions on a cheap side network, then send a summary or a proof of them down to Layer 1. The user pays a fraction of the fee while security still rests on the parent network.

The cost of that shows up on the way out. Withdrawing funds from some of these layers back to Layer 1 can take days, because of a challenge window that allows fraud to be caught. A product team needs to know that number exactly before promising a customer an instant withdrawal.

Finality is not yes or no

The concept that matters most to payments here is finality (النهائية): the point at which a transaction can no longer be undone.

On proof-of-work networks like Bitcoin, finality is probabilistic. A transaction does not become certain at one instant. The probability of it being reversed simply falls as more blocks are built on top. Hence the convention of waiting for a number of confirmations before releasing goods.

On many proof-of-stake networks, finality is deterministic: after a defined period the transaction reaches a point the network treats as conclusively final.

The difference is not academic. A treasury team that needs to know the moment after which it can book an entry without qualification treats the two cases completely differently. Compare this with sarie, which gives you known instant finality inside a clear domestic framework.

Bridges: the most dangerous piece

A bridge (الجسر) moves value from one network to another. It works by locking the asset on the first network and issuing a representation of it on the second. Which means the bridge holds real money locked up, and that makes it an ideal target.

This is not theoretical. Around 625 million dollars was stolen from the Ronin bridge in March 2022, around 325 million from Wormhole in February 2022, and around 611 million from Poly Network in 2021. The largest incidents in this field hit bridges, not individual wallets.

Treat a bridge like a correspondent bank with its worst failure modes. Decide in advance what you tell a customer whose money is stuck, how long you wait before escalating, and who absorbs the loss. Banks have answered these questions about SWIFT outages for decades. They are the same questions here with new names.

Oracles

An oracle (الأوراكل) is what brings outside-world data into smart contracts: a price, an index, the outcome of an event. Chainlink is the best-known provider.

Do not treat it as a technical detail. A lending protocol relying on a price from a single manipulable source will liquidate healthy positions or issue loans against phantom collateral. Oracle choice therefore belongs in risk design, and gets questioned the way the source of any sensitive financial data does.

How a product team chooses

Start from the use case, not the network. Small repeated payments need a low and predictable fee. Large treasury settlement can carry a higher fee but demands clear finality and institutional custody tooling. Tokenising an asset needs a network the custodian and the auditor will accept before the developer does.

Then ask about liquidity rather than brand. The same token name existing on three networks does not mean its liquidity is unified underneath, and you may find a deep market on one and almost none on another.

Then test failure before launch. What if the network halts for an hour, if a Layer 2 withdrawal takes a week, if the bridge freezes. A serious product is recognised by its answer to those three, not by its speed on a good day.