EXIL-CBI study materials
EXIL-CBI study resources

EXIL-CBI practice questions

One question at a time. Build your EXIL-CBI understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Foundations and Terminology

Which feature most clearly distinguishes a blockchain ledger from a conventional centralized database?

  1. Every record is encrypted with a symmetric key
  2. Only the operator may read the stored records
  3. It is replicated and append-only across nodes Correct answer.
  4. All records are kept on a single authoritative server

Explanation

A blockchain's defining trait is that many nodes hold and extend the same append-only ledger, so no single party controls the record. That combination of replication and append-only structure is what replaces a central authority.

Question 2Foundations and Terminology

Which property of a cryptographic hash function guarantees that hashing the same message twice produces an identical digest?

  1. Collision resistance
  2. The avalanche effect
  3. Preimage resistance
  4. Determinism Correct answer.

Explanation

Determinism means the same input always yields the same output digest, which is what allows independent nodes to agree on a value. The other properties describe different security guarantees.

Question 3Foundations and Terminology

What is the primary purpose of a Merkle tree in a block?

  1. Efficient proof that a transaction is in a block Correct answer.
  2. Distributing mining rewards among pool members
  3. Encrypting the payload of every transaction in the block
  4. Compressing block headers before network broadcast

Explanation

A block header commits to a Merkle root built from hashes of all transactions, so a short branch of hashes proves inclusion without transmitting the whole block. The tree is a commitment structure rather than an encryption or reward mechanism.

Question 4Foundations and Terminology

Which element of an asymmetric key pair is published so that others can send funds to the owner?

  1. The public key Correct answer.
  2. The private key
  3. The seed phrase
  4. The session key

Explanation

Funds are sent to an address derived from the public key, so the public component can be shared freely while the private key stays secret. Only the private key can authorise spending.

Question 5Foundations and Terminology

The first block in a blockchain, which has no predecessor, is known as the:

  1. Genesis block Correct answer.
  2. Orphan block
  3. Uncle block
  4. Checkpoint block

Explanation

The genesis block is hard-coded as the root of the chain and contains no parent hash, and every later block traces back to it.

Question 6Foundations and Terminology

A node that downloads and independently validates the entire blockchain history is called a:

  1. Mining pool
  2. Full node Correct answer.
  3. Relay node
  4. Light client

Explanation

A full node stores the complete chain and checks every transaction and block against the consensus rules itself rather than trusting a third party; clients that fetch only headers depend on such nodes.

Question 7Foundations and Terminology

In most public blockchains, an address is best described as:

  1. A shortened derivative of a public key Correct answer.
  2. The private key in encoded form
  3. A permanent identity verified by a government
  4. A random number stored only by the sender

Explanation

An address is produced by hashing and encoding a public key, so it can be published while the private key remains secret. It carries no legal identity and must be known to the recipient.

Question 8Foundations and Terminology

A transaction is said to have one confirmation when:

  1. It is included in a mined block Correct answer.
  2. It has been broadcast to the mempool
  3. It has been signed by the sender's private key
  4. The receiving wallet has displayed the amount

Explanation

The first confirmation is inclusion of the transaction in a block, and each later block adds one more confirmation. Signing and broadcast are earlier, unconfirmed stages.

Question 9Foundations and Terminology

Under the UTXO model, a wallet balance is determined by:

  1. Summing the unspent outputs the wallet can unlock Correct answer.
  2. Averaging the values of the wallet's recent inputs
  3. Reading the account's stored balance field in the state trie
  4. Summing all transactions ever sent to the wallet's address

Explanation

In the UTXO model coins exist as discrete outputs, so the spendable balance is the total of outputs the wallet holds keys for that no other transaction has spent.

Question 10Foundations and Terminology

A protocol change that makes previously valid blocks invalid and requires every node to upgrade is called a:

  1. Side chain
  2. Soft fork
  3. Hard fork Correct answer.
  4. State channel

Explanation

A hard fork changes the consensus rules so that blocks produced under the old rules are rejected, and nodes must upgrade or remain on a separate chain. Changes that stay compatible are soft forks.

Question 11Foundations and Terminology

In proof-of-work consensus, the right to add the next block goes to:

  1. The node holding the largest number of tokens
  2. The miner that first finds a qualifying hash Correct answer.
  3. The validator with the longest uninterrupted uptime
  4. The participant chosen by round-robin rotation

Explanation

Proof of work awards block production to whoever finds a nonce whose block hash satisfies the difficulty target, which proves expenditure of computational effort. Token holdings and rotation play no part in this mechanism.

Question 12Foundations and Terminology

An attacker who controls a majority of a proof-of-work network's hash rate can most directly:

  1. Decrypt transactions that were previously encrypted
  2. Steal private keys from dormant wallets
  3. Change the supply limit enforced by every wallet
  4. Reverse recent transactions by re-mining blocks Correct answer.

Explanation

A majority of hash power allows an attacker to mine a competing chain and reorganise recent history, enabling double spends. Signatures still cannot be forged, so keys and encrypted data remain safe.

Question 13Foundations and Terminology

A smart contract is best described as:

  1. A service that converts tokens into fiat currency
  2. A legally binding PDF stored on a website
  3. Code deployed to a blockchain that runs as programmed Correct answer.
  4. A hardware wallet that signs transactions automatically

Explanation

A smart contract is executable code stored at an address on the chain whose functions run deterministically when transactions call them. It is neither a legal document nor a signing device.

Question 14Foundations and Terminology

On an account-based chain that charges gas, the fee for a transaction is calculated from:

  1. The number of confirmations received
  2. Gas consumed multiplied by the gas price Correct answer.
  3. The size of the recipient's wallet balance
  4. A fixed amount set by the receiver's address

Explanation

Fees equal the units of gas consumed by the computation multiplied by the gas price the sender offers. Confirmation count and balances do not enter the calculation.

Question 15Foundations and Terminology

A wallet whose private keys are generated and stored on a device kept offline is called a:

  1. Cold wallet Correct answer.
  2. Hot wallet
  3. Custodial wallet
  4. Watch-only wallet

Explanation

A cold wallet keeps key material on a device that never connects to the internet, which reduces the risk of remote theft. Custody describes who controls the keys, not whether the device is offline.

Question 16Foundations and Terminology

The mempool of a node holds:

  1. Valid transactions waiting to be included in a block Correct answer.
  2. A permanent archive of every historical block
  3. Blocks that failed consensus validation
  4. The private keys of the node's own wallets

Explanation

The mempool is a holding area for transactions a node has validated but that no block has yet included. Its contents differ between nodes and are not permanent.

Question 17Foundations and Terminology

Public blockchains are usually described as pseudonymous because:

  1. Transaction amounts are hidden from all observers
  2. Users must register an alias with a central authority
  3. Addresses are not tied to real-world identities by design Correct answer.
  4. Every address is publicly linked to a verified name

Explanation

Addresses act as pseudonyms: they are not registered identities, yet every transaction is public, so linkage to a person is possible using additional data. That differs from true anonymity.

Question 18Foundations and Terminology

Which statement best distinguishes a token from a native coin?

  1. A token always requires its own separate blockchain
  2. A token is issued on top of an existing blockchain Correct answer.
  3. A token cannot be transferred between addresses
  4. A token is only usable inside a single wallet

Explanation

Tokens are created by contracts deployed on an existing chain and inherit its security, whereas a native coin is the base asset of its own chain. Both can be transferred between addresses.

Question 19Foundations and Terminology

A blockchain oracle is used to:

  1. Mine blocks on behalf of a consortium
  2. Supply external data to on-chain contracts Correct answer.
  3. Recover lost private keys for users
  4. Split a chain into two competing forks

Explanation

Oracles feed off-chain information such as price or event data into contracts, which cannot read the outside world on their own. They do not produce blocks, recover keys, or alter consensus.

Question 20Foundations and Terminology

A Layer 2 network is designed primarily to:

  1. Replace the underlying blockchain entirely
  2. Provide legal identity for on-chain accounts
  3. Store private keys in a hardware module
  4. Increase throughput while relying on a base chain Correct answer.

Explanation

Layer 2 systems move execution off the base chain while inheriting its security, which raises throughput and lowers cost. They complement the base layer rather than replacing it.

Question 21Foundations and Terminology

Which statement best describes a soft fork?

  1. It splits the network into two permanently incompatible chains
  2. It requires every node to delete and resync the chain
  3. It tightens rules so old nodes still accept new blocks Correct answer.
  4. It changes the block reward without node upgrades

Explanation

A soft fork restricts the rule set, so blocks valid under the new rules remain valid to unupgraded nodes and the chain does not split if adoption is sufficient. A hard fork is the incompatible case.

Question 22Foundations and Terminology

Which statement about finality is accurate?

  1. Finality is guaranteed by the wallet software alone
  2. It strengthens as more blocks are built on top Correct answer.
  3. Finality means the private key can no longer be used
  4. A transaction is final once it enters the mempool

Explanation

In probabilistic systems, confidence grows with each block built on the block containing the transaction, and some protocols add explicit finality checkpoints. Finality is a consensus property, not a wallet feature.

Question 23Foundations and Terminology

A verifier holds only a block header and a Merkle branch for a transaction, and recomputing the branch hashes yields exactly the header's Merkle root. What has been proven?

  1. The transaction output has not been spent
  2. The transaction has at least six confirmations
  3. The transaction sender owned the private key
  4. The transaction is included in that block Correct answer.

Explanation

A Merkle branch lets a verifier recompute the root from one leaf and its sibling hashes, and matching the header root proves that leaf was committed to in that block. It says nothing about signatures, spends, or confirmation depth.

Question 24Foundations and Terminology

A Bitcoin-style transaction has two outputs: one pays a previously unseen address whose script type matches the inputs, for an odd amount. An investigator would most likely treat that output as:

  1. A payment to a third-party merchant
  2. Change returned to the sender Correct answer.
  3. A miner fee calculation error
  4. A burn to an unspendable address

Explanation

A newly created output whose script type matches the inputs and whose value is not round usually indicates the sender paying themselves change. Genuine payments more often use round amounts and addresses with prior history.

Question 25Foundations and Terminology

A chain reorganisation occurs when:

  1. Nodes adopt a competing chain with more total work Correct answer.
  2. A wallet derives a new address from its seed
  3. Two nodes disagree about the current mempool contents
  4. A block producer edits a confirmed block in place

Explanation

A reorganisation happens when nodes switch to a competing branch that has accumulated more work, abandoning blocks that were previously on the active chain. Confirmed blocks change only by replacing the branch, never by in-place editing.

Question 26Foundations and Terminology

The very first block in a blockchain, which has no predecessor block, is called the:

  1. Orphan block
  2. Coinbase block
  3. Genesis block Correct answer.
  4. Checkpoint block

Explanation

A genesis block is the initial block of a chain and is typically hardcoded into the node software, with its previous-block reference set to a null or zero value. Every later block traces its ancestry back to it, so all nodes agree on a single starting point. Blocks that are valid but left out of the main chain, trusted reference points, and the first transaction in a block are separate concepts.

Question 27Foundations and Terminology

Which property of cryptographic hash functions makes it impractical to alter a past block without other nodes noticing?

  1. They can be reversed with the correct key, letting nodes recover the original data
  2. They encrypt block contents so that only block producers can read them
  3. They compress any input to a fixed size, so small differences are discarded
  4. A small change to the input produces an unpredictable, very different output Correct answer.

Explanation

Hash functions are designed so that changing even one bit of input yields an output that looks unrelated to the original hash. Because each block header commits to the hash of the previous block, altering history changes every subsequent hash and is immediately visible to anyone recomputing them. Hash functions are one-way and are not used to encrypt block data.

Question 28Foundations and Terminology

The property that makes it costly for one party to create many fake identities and gain outsized influence over a public blockchain is called:

  1. Liveness
  2. Sybil resistance Correct answer.
  3. Atomicity
  4. Finality

Explanation

Sybil resistance means that identities are not free: each unit of influence must be backed by a scarce resource such as computational work or staked capital. This stops an attacker from multiplying cheap pseudonymous identities to dominate block production or voting. Finality, liveness and atomicity describe other guarantees about transaction settlement and system behaviour.

Question 29Foundations and Terminology

Which statement best distinguishes a permissioned blockchain from a permissionless one?

  1. A permissioned chain restricts who may validate blocks or submit transactions. Correct answer.
  2. A permissioned chain must be open to the public so that any user can read its contents.
  3. A permissioned chain does not use cryptography, so blocks are not linked together by hashes.
  4. A permissioned chain has no consensus mechanism because all participants are trusted in advance.

Explanation

The defining feature is gated participation: an operator or consortium decides which parties may run validators or write to the ledger. Permissionless chains instead let anyone join, read, and compete to produce blocks, relying on economic incentives for security. Permissioned chains still use hashing, still run a consensus protocol, and may or may not be publicly readable.

Question 30Foundations and Terminology

On an account-based chain, each transaction from an account carries a strictly increasing nonce. Which problem does this mainly solve?

  1. Replaying a signed transaction a second time Correct answer.
  2. Spending an account's balance twice across two different transactions
  3. Paying transaction fees to the block producer in the right currency
  4. Compressing many transactions from the same account into one block entry

Explanation

The nonce is part of the signed transaction data and must match the account's current expected value. Once a transaction is included, its nonce is consumed, so re-broadcasting the identical signed transaction is rejected as a duplicate. This also fixes the ordering of an account's transactions.

Question 31Core Concepts and Principles

Which statement best describes the primary purpose of business intelligence?

  1. To store raw data exactly as captured by source applications
  2. To automate the correction of errors in source systems
  3. To replace the transactional systems used for daily order entry
  4. To support decisions by turning data into actionable insight Correct answer.

Explanation

Business intelligence combines data, tooling and process to convert raw records into information that supports analysis and decision-making. Its output is insight and reporting rather than the capture of transactions. Source-system correction remains an operational responsibility.

Question 32Core Concepts and Principles

A system that captures individual sales orders as they happen and must respond in under a second per transaction is best described as:

  1. An OLAP cube
  2. An OLTP system Correct answer.
  3. A data warehouse
  4. A staging area

Explanation

OLTP systems are optimized for high volumes of short, single-record read and write transactions. Analytical structures such as cubes and warehouses are instead tuned for large scans and aggregation. A staging area is a transient loading zone, not a transaction-processing platform.

Question 33Core Concepts and Principles

In a dimensional model, which content does a fact table hold?

  1. Hierarchical relationships among reporting levels
  2. Descriptive attributes used for filtering and grouping
  3. Numeric measurements of business events, keyed to dimensions Correct answer.
  4. Business rules governing data quality

Explanation

A fact table stores the measurable results of business events together with foreign keys that reference the dimensions. Descriptive context lives in the dimension tables, not in the fact table. Business rules are implemented in ETL logic or the semantic layer.

Question 34Core Concepts and Principles

What is the main role of a dimension table?

  1. To store numeric measures aggregated across events
  2. To enforce referential integrity between source systems
  3. To provide the descriptive context by which facts are filtered and grouped Correct answer.
  4. To hold temporary data during the load process

Explanation

Dimensions supply the descriptive attributes, such as product, customer or date, that give meaning to facts and act as the basis for filtering and grouping. Measures belong in the fact table. Referential integrity between source systems is not a warehouse modelling concern.

Question 35Core Concepts and Principles

Which sequence describes the three stages of an ETL process?

  1. Explore, test, launch
  2. Extract, transfer, link
  3. Extract, transform, load Correct answer.
  4. Encode, translate, load

Explanation

ETL stands for extract, transform and load: data is pulled from sources, cleaned and restructured, then written into the target warehouse. The other sequences name unrelated project or coding activities.

Question 36Core Concepts and Principles

A key performance indicator is best characterized as:

  1. A measure tracked against a target Correct answer.
  2. The primary key of a dimension table
  3. Any field used as a dashboard filter
  4. A raw transaction record from a point of sale

Explanation

A KPI is a measure tied to a target so that progress against a business objective can be judged. Without a target or threshold, a number remains an ordinary metric. Filters, keys and raw records carry no performance target.

Question 37Core Concepts and Principles

Which structure describes a star schema?

  1. A normalized chain of dimension tables with no fact table
  2. A single flat table holding all measures and attributes
  3. A central fact table joined to a set of denormalized dimension tables Correct answer.
  4. A set of tables linked only by natural keys from source systems

Explanation

A star schema places one fact table at the centre with a single join path to each surrounding dimension. The dimensions are deliberately denormalized so that queries need few joins. Normalized chains of dimensions describe a snowflake, not a star.

Question 38Core Concepts and Principles

Which property distinguishes a data warehouse from an operational system?

  1. It is optimized for high-volume single-row updates
  2. It stores only the transactions of the current day
  3. It has no defined data model
  4. It is subject-oriented and integrated across source systems Correct answer.

Explanation

Warehouses are built around business subjects and integrate data from many source systems into consistent definitions. They are non-volatile and hold historical data rather than only the current day. High-volume single-row update performance is a hallmark of operational systems.

Question 39Core Concepts and Principles

A fact table records one row per order line, and users need totals per order. How should the order-level total be produced?

  1. By adding an order-level key to the product dimension
  2. By summing the line-level measures up to the order level Correct answer.
  3. By removing the product dimension from the model
  4. By capturing the total in the source system before extraction

Explanation

The fact table grain is the order line, so a coarser order-level total is obtained by aggregating the additive measures over the order. Adding keys or dropping dimensions does not change the grain. Pre-computing totals in the source simply duplicates data the warehouse can derive.

Question 40Core Concepts and Principles

A customer moves from Gold to Platinum status and the previous status must remain queryable. Which technique fits?

  1. Type 2 new row with a new key and validity dates Correct answer.
  2. Type 3 extra column holding only the prior status
  3. Deletion of the superseded customer record
  4. Type 1 overwrite of the status attribute

Explanation

A Type 2 slowly changing dimension inserts a new version of the row with its own surrogate key and effective dates. This preserves unlimited history and lets facts join to the version that was current at the time. A Type 1 overwrite destroys history, and deletion destroys it entirely.

Question 41Core Concepts and Principles

What is the principal trade-off of snowflaking a dimension?

  1. Loss of the ability to track history
  2. Less redundancy but more joins at query time Correct answer.
  3. Inability to use surrogate keys
  4. Faster queries but duplicated descriptive data

Explanation

Snowflaking normalizes a dimension into related tables, which removes repeated descriptive values but forces extra joins. Star schemas instead accept redundancy to keep queries simple and fast. History tracking and surrogate keys are unaffected by this choice.

Question 42Core Concepts and Principles

A user moves from a yearly revenue view to a monthly revenue view of the same data. This interaction is called:

  1. Pivoting
  2. Rotating
  3. Drilling down Correct answer.
  4. Slicing

Explanation

Drilling down navigates from a coarser level of a hierarchy to a finer level, such as year to month. Rolling up would do the reverse. Slicing, pivoting and rotating change the selection or orientation of dimensions rather than the level of detail.

Question 43Core Concepts and Principles

Two separate data marts report sales using an identical definition of customer. Which design element makes this consistency possible?

  1. A separate customer dimension in each mart
  2. A degenerate dimension in the fact table
  3. A junk dimension that combines several status flags
  4. A conformed dimension shared by both marts Correct answer.

Explanation

A conformed dimension is a single, shared dimension table with agreed keys and attributes that multiple fact tables or marts reuse. It guarantees that the same customer means the same thing everywhere. Independent dimensions in each mart would drift apart in definition and content.

Question 44Core Concepts and Principles

Why is a surrogate key assigned in a dimension table?

  1. To enforce a many-to-many relationship with the fact table
  2. To encrypt the natural key for security reasons
  3. To provide a stable, compact identifier that insulates the warehouse from source key changes Correct answer.
  4. To store the business meaning of each attribute

Explanation

A surrogate key is a warehouse-generated integer that has no business meaning, so it stays stable even if the source key or business rules change. It also keeps joins narrow and supports multiple versions of the same natural key. Security, semantics and cardinality are not its purpose.

Question 45Core Concepts and Principles

How does a dependent data mart differ from an independent data mart?

  1. It is always built in a spreadsheet
  2. It is fed from the enterprise data warehouse Correct answer.
  3. It contains no dimension tables
  4. It must be implemented as a snowflake schema

Explanation

A dependent data mart sources its data from a central enterprise data warehouse, so it inherits consistent definitions. An independent data mart draws directly from operational sources and risks divergence. Schema style and structure are unrelated to the distinction.

Question 46Core Concepts and Principles

Restricting a cube to one product category and then examining revenue by quarter for two selected regions is an example of:

  1. Roll-up
  2. Pivot
  3. Drill-across
  4. Slice and dice Correct answer.

Explanation

Slicing fixes a dimension to a single value, and dicing selects a sub-cube across several dimensions. Together they narrow the data set being viewed. Roll-up aggregates to a coarser level, while drill-across combines data from multiple fact tables.

Question 47Core Concepts and Principles

Which item belongs in a fact table as a measure rather than as a dimension attribute?

  1. Product category name
  2. Extended sales amount Correct answer.
  3. Store city
  4. Order date

Explanation

Extended sales amount is a numeric value that can be summed across business events, which makes it a measure. Category, city and date are descriptive attributes that provide context and belong in dimensions. Dates are represented in the fact table as a foreign key, not as a measure.

Question 48Core Concepts and Principles

A model that estimates next quarter's demand from historical sales patterns is an example of:

  1. Descriptive analytics
  2. Predictive analytics Correct answer.
  3. Prescriptive analytics
  4. Diagnostic analytics

Explanation

Predictive analytics uses historical data and statistical or machine-learning models to estimate future outcomes. Descriptive analytics summarises what happened, and diagnostic analytics explains why. Prescriptive analytics goes further by recommending actions.

Question 49Core Concepts and Principles

Which requirement most strongly justifies a separate analytical environment instead of querying the operational database directly?

  1. Historical cross-functional queries must not slow transactions Correct answer.
  2. Referential integrity must be enforced on new orders
  3. Single-row lookups must complete in sub-second time
  4. Each order must be captured the moment it is entered

Explanation

Long-running analytical scans compete with transaction processing for the same resources, so they are moved to a dedicated warehouse. Order capture, integrity enforcement and fast single-row retrieval are operational requirements that argue for keeping the transactional system. They do not motivate analytical separation.

Question 50Core Concepts and Principles

Who should be accountable for the definition of a business metric such as 'active customer'?

  1. The ETL developer alone
  2. The database administrator alone
  3. A designated business data steward for that domain Correct answer.
  4. Any analyst who happens to use the metric

Explanation

Metric definitions are a governance responsibility: a named business steward for the domain owns the meaning, while technical roles implement it. Leaving definitions to technical staff or to individual analysts produces conflicting numbers. Accountability must therefore sit with the business owner.