CoStudy

HomeCertificationsSnowflake SnowPro Core › Virtual Warehouses and Compute

Virtual Warehouses and Compute — Snowflake SnowPro Core practice questions

25 multiple-choice questions and 11 flashcards on Virtual Warehouses and Compute, about 8% of the Snowflake SnowPro Core bank. Every one carries a written rationale.

Written and maintained by Nick Burton · last updated 2026-08-22 · how we write and review questions

What this chapter covers

Virtual Warehouses and Compute is one of 9 chapters in CoStudy's Snowflake SnowPro Core bank, and it holds 25 of the bank's 301 multiple-choice questions — roughly 8% of the total. That proportion is not arbitrary: chapters follow the certifying body's published exam outline, and the number of questions in each is set by that domain's published weight, so the share of your practice time this chapter takes matches the share of the real exam it accounts for.

Studying by chapter is worth doing once you have a diagnostic score. A single overall percentage tells you whether you are close; it does not tell you which domain is dragging. Working a weak chapter in isolation, and re-testing it in isolation, is the fastest way to move a score that has stalled — and it is why the mock exams in CoStudy report by domain rather than as one number.

Free Virtual Warehouses and Compute practice questions

10 questions drawn from this chapter, with the full rationale shown — the controlling principle behind the right answer, and why each wrong option tempts and fails.

What does Query Acceleration Service do?

  1. Automatically speeds up every query in the account
  2. Replaces virtual warehouses with serverless compute
  3. Compresses query result sets before returning them
  4. Offloads parts of scan-heavy queries to serverless

Answer: D — Offloads parts of scan-heavy queries to serverless

D) Correct — Query Acceleration Service moves portions of eligible long-running, scan-heavy queries onto serverless compute. A) Only queries that qualify are accelerated. B) It supplements a warehouse rather than replacing it. C) It does not compress results.

A single huge query that scans a TB and joins many tables is slow. Best action?

  1. Enable multi-cluster with a MAX_CLUSTER_COUNT of four
  2. Increase the warehouse size to add memory and cores
  3. Add more concurrent users to spread the work around
  4. Disable the result cache before running the query

Answer: B — Increase the warehouse size to add memory and cores

B) Correct — scaling up gives a single query more memory and more cores, which is what a large scan-and-join needs. A) Multi-cluster addresses concurrency, not single-query speed. C) More users worsen contention. D) Disabling the cache helps nothing.

A multi-cluster warehouse serves a BI dashboard where users tolerate a few seconds of queuing but the team wants to minimize credit burn. Which scaling policy setting is MOST appropriate?

  1. Standard, which starts an additional cluster as soon as queuing is detected
  2. Economy, which favors fuller clusters and accepts some additional queuing
  3. Maximized mode, with minimum and maximum cluster counts set equal
  4. Standard with auto-suspend disabled so clusters stay warm all day

Answer: B — Economy, which favors fuller clusters and accepts some additional queuing

B) Correct — the Economy policy waits until there is enough load to keep a new cluster busy, conserving credits at the cost of some queuing, which this team accepts. A) is the right answer to the opposite question, where minimizing queuing matters most. C) pins every cluster on continuously and is the most expensive configuration available. D) removes suspension entirely and increases idle spend rather than reducing it.

Increasing warehouse size from Medium to Large does which?

  1. Doubles credits/hour and doubles available compute power
  2. Halves credits/hour while keeping the same compute power
  3. Adds identical clusters instead of larger compute nodes
  4. Triggers replication of the warehouse to another region

Answer: A — Doubles credits/hour and doubles available compute power

A) Correct — each size step doubles credits/hour (M=4 → L=8) and doubles compute. B) Inverse of the actual billing effect. C) Adding clusters is multi-cluster, a separate setting. D) Replication is unrelated to warehouse size.

A warehouse has AUTO_RESUME set to FALSE and is currently suspended. A user submits a query against it. What happens?

  1. The query is routed to another running warehouse in the same account
  2. The warehouse resumes anyway, since auto-resume applies only to multi-cluster
  3. The query fails or waits until someone resumes the warehouse manually
  4. Cloud services executes the query without compute at reduced performance

Answer: C — The query fails or waits until someone resumes the warehouse manually

C) Correct — with auto-resume off, a suspended warehouse must be resumed explicitly before it will serve queries. B) invents a multi-cluster exemption that does not exist. A) is a misconception: Snowflake never silently reroutes work to a different warehouse. D) is wrong because cloud services does not execute scans, though it can serve a cached result for an exact repeat.

Multi-cluster SCALING_POLICY = ECONOMY (vs STANDARD) means:

  1. Scale out sooner so that queries never wait in a queue
  2. Auto-suspend is turned off for every cluster involved
  3. The warehouse is limited to a single running cluster
  4. Scale out conservatively, allowing queuing to cut cost

Answer: D — Scale out conservatively, allowing queuing to cut cost

D) Correct — ECONOMY favors credit savings over latency, letting queries queue before starting another cluster. A) Aggressive scale-out is the STANDARD policy. B) Auto-suspend is configured separately. C) The warehouse still runs multiple clusters.

If many queries show 'remote disk I/O' as a big stage, the fix is usually:

  1. Add more concurrent users so that the cache stays warm
  2. Disable the result cache so reads go straight to storage
  3. Resize the warehouse, or rewrite the query to scan less
  4. Move the account to a different cloud provider region

Answer: C — Resize the warehouse, or rewrite the query to scan less

C) Correct — a larger warehouse provides more local SSD cache, and filtering more aggressively reduces the remote reads outright. A) More users increase contention. B) The result cache reduces work rather than causing it. D) The cloud provider is rarely the cause.

Which statement BEST describes how virtual warehouse credits are consumed?

  1. Credits accrue per query, in proportion to bytes scanned by that query
  2. Credits accrue by the hour, so a short burst still bills a full hour
  3. Credits accrue while the warehouse runs, with a minimum charge on each resume
  4. Credits accrue only during the seconds a query is actively executing on it

Answer: C — Credits accrue while the warehouse runs, with a minimum charge on each resume

C) Correct — billing tracks the time the warehouse is running, with a short minimum charge each time it resumes and finer-grained billing after that. A) describes a serverless per-scan model, not warehouse credits. B) is the outdated full-hour rounding belief. D) is the tempting near-miss: an idle but running warehouse still consumes credits until it suspends.

WAREHOUSE_LOAD_HISTORY in ACCOUNT_USAGE shows:

  1. Successful and failed login attempts for each user
  2. Warehouse utilization: running and queued queries over time
  3. Bytes of storage consumed by tables and internal stages
  4. Replication lag between primary and secondary databases

Answer: B — Warehouse utilization: running and queued queries over time

B) Correct — WAREHOUSE_LOAD_HISTORY reports running and queued load over time, which reveals undersized warehouses and queuing patterns. A) That is LOGIN_HISTORY. C) Storage is reported in the storage usage views. D) Replication has its own views.

Setting a multi-cluster warehouse's MIN_CLUSTER_COUNT equal to its MAX_CLUSTER_COUNT produces which behavior?

  1. Auto-scale mode, where clusters start and stop with demand
  2. The warehouse refuses to start until the counts differ
  3. Maximized mode, where all clusters run whenever the warehouse is running
  4. Clusters start on demand but never shut down until suspend

Answer: C — Maximized mode, where all clusters run whenever the warehouse is running

C) Correct — equal minimum and maximum counts put the warehouse in maximized mode, so every cluster runs while the warehouse is resumed. A) describes what happens when the minimum is lower than the maximum. B) invents a validation error that does not exist. D) is a plausible-sounding hybrid, but maximized mode starts all clusters immediately rather than on demand.

Virtual Warehouses and Compute flashcards

4 cards from the 11 in this chapter.

Virtual Warehouse?

MPP compute cluster on demand. Sizes from X-Small (1 credit/hr) to 6X-Large. Pay only when running.

What is the difference between maximized mode and auto-scale mode for a multi-cluster warehouse?

Maximized mode: minimum and maximum cluster counts are set to the same value greater than one, so all clusters start immediately and stay running. Auto-scale mode: minimum is lower than maximum, so Snowflake starts and stops clusters between those bounds according to the scaling policy.

Credits?

Snowflake's unit of compute billing. Consumed by warehouses, serverless features, cloud services (limited free portion).

What are the two multi-cluster scaling policies and how do they differ?

Standard — favors starting clusters quickly to minimize queuing, adding a cluster as soon as a query queues. Economy — favors keeping clusters fully loaded to conserve credits, only starting a cluster when there is enough queued work to keep it busy for several minutes, so some queuing is accepted.

Practise the full chapter

These are a sample. The full Virtual Warehouses and Compute chapter runs 36 items with per-chapter progress tracking, on the web and in the iOS app.

Open Snowflake SnowPro Core in CoStudy →

Other Snowflake SnowPro Core chapters

All Snowflake SnowPro Core practice questions →