Blockchain infrastructure, answered.

A consensus mechanism is the rule set that lets a decentralized network of computers agree on the order of transactions. No central authority makes that call. Proof of Work, used by Bitcoin, and Proof of Stake, used by Ethereum, Solana and Cardano, are the two dominant families. Proof of Work has computers, called miners, compete to solve a computational puzzle. Whoever solves it first gets to add the next block and earn a reward for the energy spent. Proof of Stake instead selects a validator based on how much of the network's own coin they have locked up, called staking. That replaces energy spent with capital committed. Both approaches solve the same core problem: stopping any single participant from rewriting transaction history to spend the same coin twice. Beyond the two dominant families, other variants exist. These include Delegated Proof of Stake, Proof of History, and Byzantine Fault Tolerant models, used mostly by permissioned chains built for enterprise use. Consider an investor holding a 5 lakh rupee position split across Bitcoin and Ethereum. Bitcoin's balance is secured by Proof of Work miners competing worldwide. Ethereum's balance is secured by Proof of Stake validators who have staked their own capital. Two different mechanisms protect two parts of the same portfolio. Switching consensus mechanisms is rare and difficult once a chain has real value locked into it. Ethereum did it in 2022, moving from Proof of Work to Proof of Stake, a multi year project known as the Merge. Most other established chains never attempt it, since coordinating that shift across a decentralized network carries real risk of splitting the chain in two. Qatobit's QSI Core, Growth and VRION indexes hold Bitcoin and Ethereum among their underlying assets. Each is secured by its own consensus mechanism working in the background. The investor holds the basket rather than running a node or a validator, and institutional custody handles the assets day to day.

Blockchain interoperability is the ability for separate blockchains to exchange data and assets directly with each other, without routing through a centralized intermediary. Without it, each chain functions as an isolated ledger that cannot natively read or verify what is happening on any other chain. Moving a stablecoin from Ethereum to Solana is a concrete example. Ethereum has no built in way to read Solana's state, and Solana has no way to read Ethereum's. The transfer needs an interoperability mechanism, commonly a bridge, that locks the asset on one chain and mints an equivalent on the other. A bridge is only one type of interoperability tool. Others include cross-chain messaging protocols that pass data rather than assets. Wrapped tokens represent one chain's asset in a form a different chain can use. Each approach adds its own trust assumptions, since something outside the two chains has to vouch for what happened on the other side. Picture an investor moving the rupee equivalent of 2 lakh rupees in a stablecoin from an Ethereum wallet to a Solana wallet. They want to use a service that only runs on Solana. They send the asset into a bridge and wait for it to lock on Ethereum and mint on Solana. Only then can they use it on the new chain. Without interoperability, liquidity and users stay siloed on whichever chain they started on, which is exactly the problem bridges exist to solve. Bridges have also been the single most exploited category of infrastructure in crypto. Locking large sums of value in one contract creates a concentrated target for attackers.

Hyperledger Fabric uses a pluggable ordering service instead of one fixed consensus algorithm. The network operator chooses which consensus implementation actually runs underneath, commonly Raft in current versions. Bitcoin and Ethereum fix their algorithm for every participant, but Fabric's ordering layer can be swapped by whoever operates the network. Raft is the supported ordering implementation in current versions. It is crash fault tolerant, meaning the network keeps functioning if some ordering nodes fail or go offline. Earlier versions of Fabric supported Kafka instead, before the project moved to Raft as the standard choice. Fabric is a permissioned blockchain, which changes what consensus even needs to solve. Only approved organizations run the ordering and peer nodes, unlike a public network like Bitcoin or Ethereum, where anyone can join. That known, vetted set of participants is why Fabric can use a simpler ordering mechanism instead of the energy intensive competition public chains rely on. An Indian enterprise consortium might build a supply chain platform on Hyperledger Fabric, tracking goods worth several crore rupees as they move between partner companies. That consortium would run Raft ordering among a small set of known, permissioned nodes. That differs sharply from opening the network to anonymous public participants the way a public crypto network does. A permissioned model trades decentralization for speed and control, which suits an enterprise consortium. That is a different tradeoff from a public blockchain investors actually hold assets on. Fabric's assets and transactions are visible only to the approved participants, not to the public the way a Bitcoin or Ethereum transaction is.

A node is a computer running client software, such as Bitcoin Core or Geth. It keeps its own copy of the ledger and relays transactions and blocks to other nodes. A block is a bounded batch of validated transactions bundled together with a header, which is the data unit the node stores. The header is what makes a block part of a chain rather than a standalone file. It carries a timestamp, a nonce, and the hash of the previous block. That hash is what links each block to the one before it, forming the unbroken sequence the whole network agrees on. Bitcoin targets one new block roughly every 10 minutes, added by whichever miner solves that round's puzzle first. A node's job differs from a miner's. It validates and stores every block it receives, checking the rules were followed, rather than creating new blocks itself. Consider an investor holding a 5 lakh rupee Bitcoin position. Their transaction, once broadcast, gets picked up by miners and bundled into a block alongside other transactions. It is confirmed roughly every 10 minutes as new blocks are added. Thousands of nodes worldwide then store and relay that same block, each independently checking it followed the rules. Running a node requires no mining or staking, and most nodes never create a single block. Anyone can run one purely to independently verify the chain, a different role from the miners or validators who actually produce new blocks. Bitcoin has tens of thousands of nodes doing exactly that.

Bitcoin uses Proof of Work as its consensus mechanism, secured by the SHA-256 hashing algorithm. Miners compete to find a valid block hash below a difficulty target. Whoever finds it first earns the right to add the next block and collect the reward. The difficulty target adjusts roughly every two weeks, tightening if blocks have been found too fast and loosening if they have been found too slow. That adjustment keeps the average time between new blocks close to 10 minutes, regardless of how much total computing power miners throw at the network. Finding a valid hash is pure trial and error. Miners run enormous numbers of guesses per second. The target's difficulty is set so that, across the whole network, a valid answer turns up roughly every 10 minutes. There is no shortcut, only raw computing power and luck. Consider an investor holding a 5 lakh rupee Bitcoin position. Every transaction moving that value gets bundled into a block once a miner successfully solves the difficulty puzzle. The more blocks that get built on top of it afterward, the more expensive it becomes for anyone to try to rewrite that history. Bitcoin has never switched consensus mechanisms since it launched in 2009, unlike Ethereum, which moved from Proof of Work to Proof of Stake in 2022. Proposals to change Bitcoin's mechanism have come up periodically, but none has gathered enough support across miners, developers and the wider network to actually happen. Qatobit's QSI Core, Growth and VRION indexes hold Bitcoin as part of their underlying construction. That exposure is secured by this same Proof of Work mechanism working in the background. The investor holds the basket through institutional custody, rather than mining or verifying blocks themselves.

Solana uses a hybrid system. Validators reach agreement through Proof of Stake, but a separate mechanism called Proof of History orders transactions first. Proof of History timestamps each transaction before consensus even starts. That ordering step is what lets Solana target thousands of transactions per second. Proof of History is a verifiable sequence of timestamps, generated by repeatedly hashing data in order. It works like a cryptographic clock that proves how much time passed between two events. Because transactions already carry a verified order, Solana's validators do not have to negotiate that order from scratch. This is why Solana is often described as high throughput proof of stake, rather than pure proof of stake. Validators must stake SOL, Solana's native token, to take part in block production. The system picks a leader validator for each time slot based on stake weight. That leader proposes the next block, and other validators then vote to confirm it. A validator that misses too many votes is marked delinquent and stops earning rewards. Consider an investor holding 5 lakh rupees worth of SOL. Every transfer of that position confirms in under a second because ordering is already settled by Proof of History. Confirmation still needs validator votes weighted by staked SOL, so the two mechanisms work together on every transaction. Proof of History only orders transactions; agreement on which chain is correct still runs through Proof of Stake. Solana's speed advantage comes from that ordering step, layered on top of an otherwise ordinary proof of stake process. Qatobit's QSI Growth and QSI VRION indices both hold Solana as part of their construction. An investor in either index gets exposure to Solana's network without running a validator or managing a wallet. The basket handles custody and rebalancing; the underlying consensus mechanism runs in the background.

Gas limit and gas price are two separate numbers in every Ethereum transaction. Gas limit caps how much computation the transaction may use, set in units. Gas price is what you pay per unit, in gwei. A plain ETH transfer has a fixed gas limit of 21,000 units. Total fee equals gas used multiplied by gas price. Gas used is often less than the gas limit, since the limit is only a ceiling. If the transaction finishes below that ceiling, the unused gas is refunded automatically. If it needs more gas than the limit allows, it fails, and the gas already spent is not refunded. Gas price itself has two parts since Ethereum's 2021 fee update. A base fee, set by the network and burned. And a priority fee, a tip paid to the validator. Setting the gas limit too low is a common mistake. The transaction then fails, and the sender still pays for the gas consumed before it failed. Take an investor moving 5 lakh rupees worth of ETH to a new wallet. At a gas price of 20 gwei and the standard 21,000 unit limit, the transaction consumes 420,000 gwei, about 0.00042 ETH. At roughly 2.8 lakh rupees per ETH, that works out to a fee of about 118 rupees. That fee holds regardless of the 5 lakh rupees being moved. A wallet suggesting too low a gas limit is the most common cause of a failed transaction. The risk is highest for a complex transaction, like a token swap. The gas already spent up to the point of failure is gone even though the transaction never completes. Most wallets now estimate the gas limit automatically, which makes this error rarer than it used to be. Qatobit investors never set a gas limit or a gas price themselves. Quick Buy/Sell and the Crypto Indices both execute in INR, with Qatobit's own transaction fee covering execution. There is no wallet to configure and no failed transaction to debug.

There is no single official count of consensus mechanisms in use today. The major families are proof of work and proof of stake. Others include delegated proof of stake, proof of history, and proof of authority. Permissioned chains add BFT style protocols on top of that list. Proof of work, used by Bitcoin, has miners compete to solve a puzzle first. Proof of stake, used by Ethereum since 2022, has validators lock up collateral instead of burning electricity. Delegated proof of stake lets token holders vote for a smaller set of block producers, trading some decentralization for speed. Proof of history, used by Solana, orders transactions before consensus runs, rather than replacing it. Proof of authority relies on a fixed set of approved validators, common on private or consortium chains. Enterprise platforms like Hyperledger Fabric use pluggable ordering services, such as Raft or BFT, instead of any public mechanism. An investor comparing where to hold 5 lakh rupees across chains is really comparing these families. A proof of work chain like Bitcoin settles slower and costs more per transaction during congestion. A proof of stake chain like Ethereum or a proof of history chain like Solana settles faster at lower cost. New consensus designs are proposed constantly, so any fixed count goes stale within a year. What matters for an investor is which family a chain uses. That determines the real tradeoff, not the total number of mechanisms that exist.

A miner and a validator both add new blocks, but they qualify differently. Miners, used by Bitcoin, compete to solve a computational puzzle first. Validators, used by Ethereum since 2022, are chosen algorithmically based on staked collateral. Running a solo validator requires a minimum of 32 ETH. In proof of work, every miner races simultaneously. Only the first to solve the puzzle earns the block reward plus transaction fees. Losing miners get nothing for that round, only the electricity they spent. In proof of stake, one validator is selected per slot to propose a block, and others attest to it. There is no computational race, so proof of stake uses far less electricity than proof of work. A validator that proposes a bad block or goes offline can be slashed, losing part of its staked ETH. Picture an investor holding 5 lakh rupees worth of ETH staked through a pool. If the pool's validator misbehaves and gets slashed, that investor's share of the staked pool absorbs a small loss. A Bitcoin miner losing a block race risks no such penalty beyond the electricity already spent. The two roles are penalized differently for failure. A losing miner only wastes electricity, a sunk cost rather than a direct loss of capital. A validator that misbehaves can lose staked ETH outright, a direct financial penalty a miner never faces.

A consensus mechanism and a smart contract sit at different layers of a blockchain. The consensus mechanism, proof of work or proof of stake, decides which transactions get included and in what order. A smart contract is application code that runs only after a transaction is already confirmed. Consensus operates at the protocol layer, common to every transaction on the chain. It decides ordering and finality, nothing about what any individual transaction does. A smart contract, written in a language like Solidity, runs on the Ethereum Virtual Machine. It defines the logic for one specific application, such as a lending pool or a token swap. The two layers can change independently of each other. Ethereum switched its entire consensus mechanism, from proof of work to proof of stake, in the 2022 Merge. Every smart contract already deployed kept running afterward, completely unaffected by that switch. Consider an investor holding 5 lakh rupees in a DeFi lending position built on a smart contract. That position's logic, interest rates and collateral rules, is unaffected by which validator proposed the block it sits in. Only if the underlying chain's consensus mechanism itself failed would the position be at risk. People often blame the blockchain itself when a smart contract bug causes a loss. The consensus mechanism usually worked exactly as designed; the bug lived in the application code layered on top of it.

Ethereum's Dencun upgrade, which went live in March 2024, cut typical Layer 2 fees by roughly 90 percent. It introduced EIP-4844, adding a new kind of cheap data space called blob space. Layer 2 networks use that space to post their transaction data back to Ethereum far more cheaply than before. Before Dencun, Layer 2 rollups had to post their data using the same expensive space as ordinary Ethereum transactions. Blob space is priced and cleared separately from that regular space, with its own base fee. That separation is what let the cost of posting data collapse without changing Ethereum mainnet's own fee market. Dencun only reduced the cost of posting Layer 2 data back to Ethereum. It did not change how gas fees work for a transaction sent directly on Ethereum mainnet. Mainnet fees still follow ordinary supply and demand for block space. Before Dencun, moving 25,000 rupees a month into an asset through a Layer 2 network cost real fees. Those fees could take a noticeable share of the amount moved. After the upgrade, that same monthly transfer typically costs a small fraction of what it did before. Layer 2 fees fell by roughly 90 percent almost overnight. Dencun's fee cut applies to the Layer 2 network's cost of posting data. It does not apply to what an app charges its own users. An app can still charge its own fee on top of a now-cheaper Layer 2 transaction.

An Ethereum gas fee tracks computational complexity. It follows the number of gas units a transaction requires, not the amount of crypto being sent. A simple transfer uses a fixed amount of gas, around 21,000 units. That holds whether it moves 100 rupees or 1 crore rupees. Gas measures computational work, not value moved. Total fee equals gas units used multiplied by gas price, paid in gwei. Sending a larger amount through the same simple transfer function consumes the exact same 21,000 units. Interacting with a smart contract uses more gas units than a plain transfer. A token swap on a decentralized exchange is one example. That is because the contract runs more code, not because more value is changing hands. A complex swap moving a small amount can cost more gas than a simple transfer moving a large one. Sending 5 lakh rupees worth of ETH in a plain transfer costs the same gas as sending 5,000 rupees worth. The two transfers use an identical function. Both use the standard 21,000 unit limit. At a given gas price, both pay the same fee in rupee terms. The confusion usually comes from comparing gas fees to a bank's percentage-based transfer fee. A gas fee is closer to a flat postage charge for the computation, not a cut of the amount transferred. Qatobit's own fee works differently from a gas fee. Quick Buy/Sell costs 0.4% of the transaction amount, and any Crypto Index basket transaction costs 0.35%. Both are shown before the investor confirms, so the cost scales with the amount rather than staying flat.

Blockchain interoperability, moving value or data between separate chains, is achieved mainly through bridges, cross-chain messaging protocols, and atomic swaps. Bridges lock an asset on its home chain and mint a wrapped version on the destination chain. LayerZero and Wormhole are the best-known messaging protocols in use today. A lock-and-mint bridge is the most common method. The original asset sits locked in a smart contract on its home chain. A wrapped token, backed 1:1, is minted on the destination chain. Cross-chain messaging protocols like LayerZero and Wormhole instead pass verified messages between chains, without necessarily wrapping a token. Some networks are built from the ground up for native cross-chain communication. Cosmos does this through its IBC protocol, instead of relying on a bolted-on bridge. An atomic swap is a third method. It is a direct trade between two chains. It either completes on both sides or fails on both, with no bridge contract in the middle. An investor moving 5 lakh rupees worth of an asset from Ethereum to another chain usually goes through a bridge. An atomic swap is used far less often for this kind of transfer. Bridges support far more asset pairs than atomic swaps do. The bridge locks the original asset and mints an equivalent wrapped token on the new chain. Every interoperability method adds a new point of failure beyond the two blockchains themselves. A bridge's smart contract, or a messaging protocol's verification set, becomes a target in its own right. That risk sits separate from either chain's own security.

Since Ethereum's 2021 EIP-1559 upgrade, a gas fee equals gas units used multiplied by base fee plus priority fee. Both fees are paid in gwei. The base fee is set by the network and burned. The priority fee is an optional tip paid to the validator. The base fee adjusts automatically block to block. It rises when the previous block was more than half full, and falls when it was less. This keeps average block usage near a target level over time. The priority fee, sometimes called a tip, goes directly to the validator that includes the transaction. Raising it during congestion is how a sender gets included faster, since validators pick the highest-tipping transactions first. Take a transaction using 100,000 gas units, a base fee of 15 gwei, and a 2 gwei tip. Total gas price is 17 gwei, for a fee of 1,700,000 gwei, about 0.0017 ETH. At roughly 2.8 lakh rupees per ETH, that works out to about 476 rupees for the transaction. The base fee is burned, permanently removed from supply, rather than paid to any validator. Only the priority fee, the tip, actually goes to whoever proposed the block. This is why a busy network can burn large amounts of ETH. Individual validators still earn comparatively little from base fees. Qatobit's Quick Buy/Sell fee skips this calculation entirely. Buying or selling a single asset costs a flat 0.4% of the transaction amount, shown before the investor confirms. There is no base fee, no priority tip, and no gwei to estimate.

There is no fixed count of Ethereum Layer 2 networks, because new ones launch regularly. L2Beat, the most widely cited independent tracker, lists dozens of active networks ranked by total value locked. Arbitrum, Optimism, Base, and zkSync are consistently among the largest by usage. Layer 2 networks execute transactions off Ethereum's main chain and post compressed data back to it. Most are built as optimistic rollups or zk-rollups, the two dominant designs in current use. The L2Beat list changes as networks launch, merge, or shut down. Any fixed number quoted today is likely to be stale within months. L2Beat also tracks how much of each network's security actually depends on Ethereum. Some rollups are more decentralized than others. That distinction matters more for risk than the raw count of networks does. A network can rank high on total value locked while still relying on a small operator set. An investor holding 5 lakh rupees across several Layer 2 networks is better served checking L2Beat's live figures. A headline count is not worth memorizing. Total value locked and the network's security model tell an investor more than how many Layer 2s exist in total. A high total value locked figure does not by itself mean a network is safe. Some Layer 2s still rely on a small, permissioned set of operators. That tradeoff is worth checking on L2Beat before treating any two networks as equivalent.

Most proof of stake chains pick the next block proposer through pseudo-random selection weighted by stake size. Ethereum specifically uses a mechanism called RANDAO, combined with each validator's effective balance. Together they assign proposer and committee duties for every slot. A slot occurs roughly every 12 seconds on Ethereum. RANDAO builds randomness by combining contributions from many validators, so no single validator can predict or control the outcome alone. The validator selected for a slot proposes the next block; a separate committee of validators then attests to it. That attestation step is what confirms the proposer actually followed the protocol's rules. Stake weighting means a validator with more staked ETH is proposed and included more often. That frequency is proportional to its share of the total stake. A validator that is selected but acts dishonestly, or goes offline when its turn comes, can be penalized through slashing. An investor with 5 lakh rupees staked through a pool owns a proportional share of that pool's stake. That share sets the pool's chance of being selected each slot. More stake means more selections over time, and more selections mean more of the block rewards the pool earns. Selection is probabilistic, not scheduled, so a small staker can go a long stretch without being chosen. That is one reason liquid staking pools exist, to smooth out reward timing across many smaller stakers. Pooling many stakers together evens out the wait between selections for each individual depositor.

An Ethereum gas fee equals gas units used multiplied by gas price in gwei. That cost is based on computation, not on value transferred. Sending 0.01 ETH and sending 100 ETH in an equally simple transfer costs the same gas. Both use the same fixed 21,000 unit limit for a plain transfer. The confusion is understandable, because most financial fees, like a card processing fee, are quoted as a percentage. Gas works more like a flat postage charge for the computation a transaction requires. A more complex transaction, like a swap on a decentralized exchange, uses more gas units than a simple transfer. That holds regardless of the value moved. That is a function of how much code runs, not of how much money is involved. Moving 5 lakh rupees worth of ETH and moving 5,000 rupees worth of ETH cost the same gas. Both use an identical simple transfer. At a given gas price, both fees come out to roughly the same rupee amount. A transaction that fails partway still burns the gas already spent, even though nothing was actually transferred. That gas cost is unrelated to the value the sender intended to move. Qatobit's own fees are percentage-based by design. Quick Buy/Sell costs 0.4% of the transaction, and a Crypto Index basket transaction costs 0.35%. Both scale with the amount invested, unlike the flat gas cost described above.

Moving to a Layer 2 network can substantially lower Ethereum gas fees. Layer 2 rollups, like Arbitrum, Optimism, and Base, execute transactions off Ethereum's main chain. They post compressed data back to it. L2 transactions commonly cost a small fraction of an equivalent mainnet transaction. A Layer 2 still charges its own fee. The cost moves to a cheaper layer rather than disappearing entirely. That fee still covers the cost of eventually posting data back to Ethereum mainnet for final settlement. Two main designs exist: optimistic rollups, like Arbitrum and Optimism, and zk-rollups, like zkSync and Starknet. Since the Dencun upgrade added cheap blob space in March 2024, that data-posting cost fell by roughly 90 percent. Combined with the L2's own lower execution cost, the total fee drops further. A user can end up paying a small fraction of a mainnet equivalent. Sending 25,000 rupees a month into an asset on Ethereum mainnet could cost several hundred rupees in gas alone. That cost is highest during a busy period. The same transfer on a Layer 2 network commonly costs a few rupees instead. Withdrawing from an optimistic rollup like Arbitrum or Optimism back to Ethereum mainnet still takes about a week. That wait is the network's challenge period. A zk-rollup avoids that delay, since its validity proof is checked mathematically rather than through a dispute window. Either way, the underlying gas cost saved by moving to Layer 2 in the first place remains.

Permissioned BFT mechanisms, used by enterprise chains like Hyperledger Fabric, are generally considered the most centralized. Only an approved, fixed set of nodes can validate a block, unlike a public network anyone can join. That approved set is often a handful of organizations, not thousands of independent participants. Among public, investable chains, decentralization is more of a spectrum than a binary label. Bitcoin's proof of work mining is spread across a globally distributed set of operators, with no membership list to approve. Thousands of independent miners compete for each block, which is what the term decentralized mining actually describes. Proof of stake chains can concentrate stake in a few large pools or exchanges. That concentration is considered more centralization-prone than Bitcoin's globally distributed mining base. A handful of large staking providers controlling most of the staked supply is the specific pattern regulators and researchers watch. An investor putting 5 lakh rupees into a proof of stake asset is trusting someone else with that stake. The trust falls on whichever validators or pools hold the largest share. Checking how concentrated that supply is tells the investor more about centralization risk than the consensus label alone. Decentralization is one leg of the widely cited blockchain trilemma, traded off against security and scalability. A chain optimized hard for speed, like a permissioned BFT network, typically gives up decentralization to get there.

In a stake-weighted proof of stake network, a validator's influence scales with how much it has staked. Ethereum requires 32 ETH to run a solo validator, which anchors that scale. More staked ETH means more chances to be selected and more rewards earned over time. This is different from one-node-one-vote; a validator staking twice as much gets roughly twice the influence over time. That design links voting power directly to economic commitment rather than to the number of participating machines. It also means voting power can shift whenever large holders move stake in or out. 32 ETH is a high bar for most individual investors. Liquid staking pools exist to combine smaller stakes into one validator's worth of influence. A few large staking pools or exchanges control a big share of total staked supply. That concentration is the main centralization concern raised about proof of stake. That concern grows or shrinks with how evenly stake is spread across providers. An investor with 5 lakh rupees staked through a liquid staking pool owns a proportional slice of the pool's stake. That slice sets their share of its combined voting power. The pool itself, not the individual investor, is the entity actually selected to propose or attest blocks. Stake-weighted voting means influence follows capital rather than headcount. A wallet holding a large share of staked supply carries outsized sway over the network. That sway holds even if the wallet represents very few individual people.

A testnet and mainnet are two separate, unconnected networks. A testnet, such as Ethereum's Sepolia, is where developers deploy and test smart contracts. They use test tokens that have no market value. Mainnet is the live production network where transactions move real assets and are treated as final once confirmed. The two networks share no common state, no shared balances, and no shared transaction history. Software and contracts are usually tested on a testnet first precisely because mistakes there cost nothing real. Ethereum has retired several older testnets over the years, most recently Goerli, and developers migrate to whichever one is current. Sepolia's test ETH is given out free by faucets and carries no value outside the testnet itself. A mainnet transaction, once confirmed, is treated as irreversible and moves real assets between real wallets. Block explorers, wallets, and RPC endpoints each point to one network or the other, never both at once. Sending 5 lakh rupees worth of real ETH requires a mainnet transaction, never a testnet one. A testnet transaction moving the equivalent amount of test ETH has no real-world value at all. Sending real assets to a testnet address will not resolve. Neither will looking up a testnet transaction hash on a mainnet explorer. The two networks share no state or value, so each system only recognizes its own transactions.

A bridge token, or wrapped token, represents an asset from one blockchain on a different blockchain. It is backed 1:1 by the original asset, which sits locked in a smart contract on its home chain. Wrapped Bitcoin, WBTC, is the best-known example, letting BTC be used inside Ethereum smart contracts. Wrapping works through a custodian or a decentralized bridge. That bridge locks the original asset and mints an equal amount of the wrapped version elsewhere. Redeeming the wrapped token burns it and releases the locked original back to the holder. The wrapped version carries its own ticker and trades as a separate asset on its new chain. The wrapped token's value depends entirely on the bridge continuing to hold the locked original asset. If that custody arrangement fails or is hacked, the wrapped token can lose its backing. This can happen even while the original chain keeps running normally. An investor holding 5 lakh rupees worth of WBTC on Ethereum is trusting the custodian. The same value in real Bitcoin must sit locked with WBTC's custodian. That trust is separate from trusting Bitcoin's own network, which never sees the wrapped token at all. A bridge token's price can briefly decouple from the original asset if the market doubts the custodian's solvency. That gap, however small, is a live signal of custody risk that the original asset itself does not carry.

A mempool, short for memory pool, is the set of transactions a node has received and validated. Each has not yet been included in a block. Each node keeps its own mempool locally. A transaction waits there until a miner or validator picks it up. Miners and validators typically pick transactions from their mempool by highest fee per unit of size first. A low-fee transaction can sit unconfirmed for hours during network congestion. A high-fee one clears within the next block instead. Because each node's mempool is separate, an unconfirmed transaction's exact position can differ slightly from node to node. Most nodes drop a transaction from their mempool after it sits unconfirmed past a set time limit. That dropped transaction never reaches the chain unless the sender rebroadcasts it with a higher fee attached. An investor sending 25,000 rupees worth of ETH with too low a fee may see it stall. The transaction can sit in the mempool for hours. That is more likely during a busy period. Raising the fee, or resending with a higher priority fee, usually gets it picked up on the next block instead. A transaction sitting in the mempool can sometimes be seen and reordered by other actors before it confirms. That practice is called front-running. This is one reason some networks and applications now offer private transaction routes that skip the public mempool entirely.

A zk-rollup bundles many transactions off-chain and submits one cryptographic validity proof, a zero-knowledge proof, with every batch. That mathematical check means a zk-rollup skips the multi-day dispute window an optimistic rollup uses. Live examples include zkSync and Starknet, both running on Ethereum today. The validity proof mathematically guarantees the batch of transactions was executed correctly, without anyone needing to check it by hand. Once that proof is verified on Ethereum, the batch is considered final almost immediately. No challenge period follows, unlike on an optimistic rollup. That is the core tradeoff between the two designs. Generating a validity proof is computationally intensive, which historically limited how complex a zk-rollup's supported smart contracts could be. That gap has narrowed as proving technology has improved. It can still make zk-rollups slower to support new contract types than optimistic rollups. Optimistic rollups accepted that risk in exchange for easier support of complex contracts from the start. Withdrawing 5 lakh rupees worth of assets from a zk-rollup back to Ethereum mainnet can complete in minutes. That speed holds once the validity proof clears. The same withdrawal from an optimistic rollup typically takes about a week instead. A third party can front the funds for a fee to skip that wait. A zk-rollup's fast, trustless finality comes at the cost of proof generation being harder to build for very complex applications. That tradeoff is narrowing as zero-knowledge proving technology matures, but it has not disappeared entirely. Choosing between the two designs often comes down to that speed-versus-flexibility tradeoff.

An optimistic rollup assumes every submitted transaction is valid by default. It only runs a fraud proof if someone disputes it. That assumption creates a challenge window, typically about 7 days, during which a disputed transaction can still be reverted. Arbitrum and Optimism are the two largest optimistic rollups by usage. No proof is required upfront, so optimistic rollups are simpler to build. They typically support more complex smart contracts than zk-rollups do. The tradeoff shows up on withdrawal, where funds must wait out the challenge window before mainnet treats them as final. Deposits into the rollup face no such delay, only withdrawals back to mainnet do. During that roughly 7-day window, anyone can submit a fraud proof against a disputed transaction. If fraud is proven, the disputed transaction is reverted and the party that submitted it is penalized. Most transactions go unchallenged and simply become final once the window closes. An investor withdrawing 5 lakh rupees worth of assets from Arbitrum back to Ethereum mainnet must wait out the challenge window. That window runs roughly 7 days. Third-party liquidity providers exist who will front the funds immediately for a fee, skipping the wait. The 7-day challenge window is the most common complaint about optimistic rollups. It delays access to funds during a withdrawal. A zk-rollup avoids that specific delay, because its validity proof is checked mathematically rather than through a dispute period.

An RPC endpoint is a URL that lets a wallet or app talk to a blockchain. It queries balances, reads contract data, and submits signed transactions. It does this without the wallet running its own full node. MetaMask and most wallets connect to a public RPC endpoint by default, from a provider chosen for them. RPC stands for remote procedure call. It lets one program ask another to run a function and return the result. For a blockchain, that function is usually reading current state or broadcasting a new signed transaction. The wallet never has to store or sync the blockchain's full history itself. Public RPC providers like Infura and Alchemy run the full nodes so individual wallets and apps do not have to. A user can swap the default endpoint for their own node's endpoint, trading convenience for more direct control. An investor checking a wallet holding 5 lakh rupees worth of ETH is really making a request. That request goes to an RPC endpoint for the wallet's current balance. Submitting a transaction to move that ETH also goes through the same RPC endpoint. If an RPC endpoint is down or rate-limited, a wallet's balance display and transaction broadcasting can fail. This can happen even though the blockchain itself is running normally. That failure looks like a network problem to the user, when it is really just one provider's endpoint having trouble.

Liquid staking lets someone earn staking rewards without locking tokens directly with a single validator. Regular node staking on Ethereum requires a minimum of 32 ETH to run a solo validator. Liquid staking pools smaller deposits together and issues a tradeable receipt token, like stETH, representing the staked position plus rewards. The pool runs the actual validators and spreads the 32 ETH requirement across many depositors. In return, each depositor receives a receipt token proportional to what they put in. The pool operator handles the technical work of running and maintaining each validator. That receipt token, unlike locked node staking, can still be used elsewhere in DeFi. The underlying stake keeps earning rewards the whole time. This is the liquidity that gives liquid staking its name. The position is not frozen for the staking period. stETH is the largest such receipt token by circulating supply today. An investor with 5 lakh rupees to stake cannot reach the 32 ETH minimum for solo staking on their own. Through a liquid staking pool, that same 5 lakh rupees earns staking rewards immediately. The receipt token stays usable elsewhere the whole time. The receipt token's price can briefly trade below the underlying staked asset's value during periods of high withdrawal demand. That gap closes over time but is a real, if usually small, risk that locked node staking does not carry.

A crypto bridge is a common hacking target. It concentrates an entire pool of locked collateral into one smart contract or a small multisig. The Ronin bridge hack in March 2022 lost about 625 million dollars after attackers compromised validator signing keys. It remains one of the largest crypto thefts on record. A bridge locks the original asset in one contract on its home chain. It mints a wrapped version on the destination chain. That means the entire backing collateral for every wrapped token in circulation sits in a single point of failure. A regular exchange or wallet hack, by comparison, usually only exposes its own users. The Wormhole bridge hack happened in February 2022 and lost about 325 million dollars. The cause was a signature verification flaw in the bridge's own smart contract. Both hacks exploited weaknesses in the bridge's own contract logic or key management. Wormhole's operator covered the loss from its own funds; Ronin's backers did the same for Ronin. An investor holding 5 lakh rupees worth of a wrapped asset is trusting the bridge's security. That trust sits on top of, and separate from, the value of the original locked collateral behind it. Checking a bridge's audit history and how its keys are managed is part of that assessment. A bridge with a small validator set is a more concentrated target than one with many independent signers. Checking how a bridge is secured matters more than checking which chains it connects.

Ethereum gas fees rise and fall because they follow demand for a fixed block space. Each Ethereum block has a gas target and a hard limit. The base fee adjusts up or down based on how full the previous block was. That adjustment is automatic, set by the protocol. When more people compete for the same limited block space, the base fee rises quickly to ration it. Demand spikes around NFT mints, major token launches, and periods of high market volatility push fees up fast. Fees are typically lower during low-activity hours, though the exact pattern shifts from week to week. Since the Dencun upgrade in March 2024, Layer 2 fees fluctuate on the same demand logic. They now move from a much lower base. An investor sending 25,000 rupees a month into ETH during a quiet period might pay a small fee. The identical transfer during a high-demand event, like a popular NFT mint, can cost several times more. The transaction itself never changes. Because the base fee adjusts automatically, there is no way to guarantee a low fee in advance. There is only a higher chance of one during historically quieter hours. Qatobit's own transaction fees do not move with network congestion. Quick Buy/Sell stays at a flat 0.4%, and a Crypto Index basket transaction stays at 0.35%, regardless of how busy the underlying network is. That fee is fixed by product, not by the moment an investor transacts.

XRP was designed as a bridge asset. It connects different currencies and assets without needing a direct trading pair between every one of them. Ripple's On-Demand Liquidity product uses XRP this way, as a temporary bridge currency for moving value across borders. The XRP Ledger settles each transaction in a few seconds. Without a bridge asset, moving value between two currencies needs a direct market between them. That direct market does not always exist at good liquidity, particularly between smaller or less-traded currency pairs. A bridge asset routes around that gap by sitting in the middle of the trade. XRP is meant to sit in the middle instead. It is converted into and then quickly back out of, in seconds, to complete the transfer. Ripple's On-Demand Liquidity product is built specifically around that bridging use case, for cross-border payment corridors. Banks and payment firms are the product's intended users, not individual traders. A business moving the equivalent of 5 lakh rupees across a currency corridor can hit a market with no direct liquidity. It can route the value through XRP instead. XRP is bought on one side and sold on the other, both legs completed within seconds. The business never has to hold XRP as a long-term position to use it this way. This bridge-currency use case is separate from XRP as a tradeable asset. People hold and speculate on XRP in its own right too. The two roles are often conflated. XRP's original design purpose was the bridging function, not being held as a long-term position.