I B M C E R T I F I C A T I O N
F1002700 IBM Certified Advanced Architect v2 PLUS IBM Professional Cloud SRE v2 Practice Exam
Exam Number: 4318 | Last updated April 17, 2026 | 395+ questions across 5 vendor-aligned objectives
The F1002700 bundle is built for senior practitioners who draft enterprise architectures and then own them through production. By joining the Advanced Architect v2 credential with Professional Cloud SRE v2, it produces rare generalists who can blueprint a platform and run it with a straight face. Candidates should be comfortable designing landing zones, engineering SLOs, and operating OpenShift or Kubernetes estates at enterprise scale.
Commanding 26% of the exam, Architecture-to-Operations Transition covers the handoff from design artifacts to runbooks, SLO registration, and capacity planning. At 22%, Reliability Engineering covers SLI/SLO design, error budgets, incident command, and chaos practice. A further 20% targets Platform Design and Operation, covering OpenShift, Kubernetes, Helm, and GitOps patterns that scale across multiple business units.
Last on the list, Observability and Incident Response accounts for 18% and spans metrics, traces, logs, IBM Instana, and PagerDuty-style incident workflows. Security and Access represents 14% and spans IAM, secrets rotation, and posture management. The exam rewards answers that show design decisions paying off in operational practice; memorize common anti-patterns and why they fail under load.
Every answer links to the source. Each explanation below includes a hyperlink to the exact IBM documentation page the question was derived from. PowerKram is the only practice platform with source-verified explanations. Learn about our methodology →
748
practice exam users
94%
satisfied users
91%
passed the exam
4.8/5
quality rating
Test your F1002700 architect v2 cloud sre v2 knowledge
10 of 395+ questions
Question #1 - Architecture-to-Operations Transition
A new enterprise platform at Carlton Rail Group is being handed from the architect to the SRE team. Runbooks are missing and the SRE lead refuses to accept the handoff.
Which handoff artifact resolves the blocker while keeping operational integrity intact?
A) A verbal walkthrough given on launch day
B) A promise to write runbooks after go-live
C) A marketing one-pager describing the platform’s features
D) A complete operational runbook covering common failures, SLI/SLO registrations, and initial capacity plan — produced before SREs accept on-call responsibility
Show solution
Correct answers: D – Explanation:
The formal handoff artifact is operational runbooks plus SLO registration plus capacity plans — the reference criteria for accepting SRE ownership. Promises, marketing docs, and verbal walkthroughs cannot be operated on-call reliably. Source: Check Source
Question #2 - Architecture-to-Operations Transition
An architect at Thornvale Biosciences drafted a sleek microservice design that the on-call SRE will reject because alert storms are inevitable at production scale.
Which adjustment makes the design operable at scale?
A) Ship the design as-is and have SREs mute alerts during incidents
B) Introduce circuit breakers, alert-grouping, and dependency-aware runbooks at design time — the ‘design for operations’ principle
C) Hand the design over and let SREs redesign it later
D) Accept the alert storms as a cost of innovation
Show solution
Correct answers: B – Explanation:
Designing for operations at design time — breakers, alert grouping, dependency-aware runbooks — is the IBM SRE reference for architecture-to-operations transitions. Muting alerts, redesigning after the fact, and accepting storms all fail the operability criterion. Source: Check Source
Question #3 - Architecture-to-Operations Transition
A platform at Westbrook Telecom must be registered with SLOs before go-live so on-call rotations know what they are defending.
Which practice satisfies the requirement?
A) Register explicit SLIs and SLOs in the observability platform (e.g., Instana) with burn-rate alerts and a named on-call rotation per SLO
B) Defer SLO definition until the first big incident
C) Let each engineer pick their own SLO informally
D) Use the architect’s gut feeling as the SLO
Show solution
Correct answers: A – Explanation:
Registered SLIs/SLOs with burn-rate alerts and owners is the IBM Cloud SRE reference for go-live readiness. Deferral and informal SLOs both defeat reliability engineering. Source: Check Source
Question #4 - Reliability Engineering
A team at Crossbow Logistics sets a 99.9% SLO. A deploy on day three of the month consumes 80% of the error budget.
Which release-cadence response aligns with error-budget policy?
A) Freeze non-reliability changes for the rest of the month and spend the remaining budget on reliability work; every exception goes through change control
B) Continue the normal release cadence because the deploy is done
C) Raise the SLO target to recover the budget
D) Fire the engineer who deployed
Show solution
Correct answers: A – Explanation:
Error-budget policy slows the release train when the budget is burning too fast. Continuing the cadence defeats the policy. Raising the target is self-deception. Firing engineers is not a reliability control. Source: Check Source
Question #5 - Reliability Engineering
A new fintech workload at Daybreak Exchange must tolerate the loss of a single availability zone with zero user impact.
Which chaos-engineering practice proves the property safely?
A) Run scheduled chaos experiments that drain a single AZ with a defined abort condition and captured metrics
B) Wait for a real AZ outage and document what happened
C) Run arbitrary kill commands in production with no safety
D) Skip the chaos practice because production is too sensitive
Show solution
Correct answers: A – Explanation:
Scheduled chaos experiments with abort conditions are the SRE practice for proving resilience claims. Waiting for outages is reactive, not proactive. Arbitrary kills are reckless. Skipping chaos leaves the resilience claim unverified. Source: Check Source
Question #6 - Platform Design and Operation
A platform architect at Summit Rail Operations must support ten business units deploying independently on one shared OpenShift estate.
Which design scales best?
A) A single namespace for every BU to share
B) One cluster per BU with no shared platform
C) A multi-tenant OpenShift design with per-BU namespaces or projects, RBAC-isolated resources, quota policies, and GitOps-managed deploys
D) Let BUs install their own Kubernetes on laptops
Show solution
Correct answers: C – Explanation:
Multi-tenant OpenShift with per-BU isolation, quotas, and GitOps is the IBM scale-out reference. Per-BU clusters multiply operational overhead without adding isolation. A single shared namespace breaks isolation. Laptop clusters are not a platform. Source: Check Source
Question #7 - Platform Design and Operation
A GitOps rollout at Stafford Harbor Freight must ensure that every environment always matches what is in Git, with drift automatically remediated.
Which capability satisfies the requirement?
A) Manual oc apply on each change
B) Argo CD configured to auto-sync and self-heal from the Git repository, with application definitions stored per environment
C) A Slack bot that announces drift without correcting it
D) Screenshots of intended state pinned to a wiki
Show solution
Correct answers: B – Explanation:
Argo CD auto-sync and self-heal is the IBM Cloud GitOps pattern for drift correction. Manual commands are drift sources, not remedies. Announcement bots and wiki screenshots do not remediate drift. Source: Check Source
Question #8 - Observability and Incident Response
During a multi-service incident at Clearwater Bank, the on-call SRE cannot identify which service triggered the cascade.
Which observability capability most directly answers ‘where did it start?’
A) Adding CPU to every service
B) Distributed traces viewed in IBM Instana, showing the first failing span and propagation across services
C) Restarting all microservices
D) Scaling down the platform until the incident resolves on its own
Show solution
Correct answers: B – Explanation:
Distributed traces identify the origin span of a cascade — IBM Instana’s canonical use. Blind CPU bumps and mass restarts do not diagnose. Scaling down is a guess, not a diagnosis. Source: Check Source
Question #9 - Observability and Incident Response
After a sev-1 at Thornridge Telecom, the team wants to capture learnings without discouraging honest disclosure.
Which practice satisfies both goals?
A) Skipping the postmortem to protect morale
B) Disciplinary action against the engineer at the keyboard
C) A blameless postmortem focused on systemic contributors and corrective actions, with findings published internally and tracked to closure
D) A closed-door review attended only by executives
Show solution
Correct answers: C – Explanation:
Blameless postmortems with tracked action items are the IBM SRE doctrine. Discipline discourages disclosure. Skipping forfeits learning. Closed-door reviews hide detail from the people closest to the problem. Source: Check Source
Question #10 - Security and Access
A platform at Rosebank Utilities grants production access to over 80 engineers, and audit finds several stale accounts.
Which IAM control best addresses the finding?
A) Give every engineer permanent cluster-admin to simplify operations
B) Share a single admin account between engineers
C) Remove IAM and rely on VPN alone
D) Switch to IAM trusted profiles and just-in-time access groups, with automated revocation on inactivity and regular access reviews
Show solution
Correct answers: D – Explanation:
Trusted profiles, JIT access groups, inactivity-based revocation, and access reviews are the IBM Cloud IAM pattern for the finding. Shared accounts break attribution. Removing IAM removes access control. Permanent cluster-admin is the opposite of least privilege. Source: Check Source
Get 395+ more questions with source-linked explanations
Every answer traces to the exact IBM documentation page — so you learn from the source, not just memorize answers.
Exam mode & learn mode · Score by objective · Updated April 17, 2026
Learn more...
What the F1002700 architect v2 cloud sre v2 exam measures
- Transition and own architecture artifacts, runbooks, SLO registration, and capacity plans to carry designs from whiteboard through production operation with full accountability
- Define and spend SLI/SLO, error budgets, incident command, and chaos practice to align engineering rhythm with business expectations for availability and pace
- Design and operate OpenShift, Kubernetes, Helm, and GitOps patterns at enterprise scale to deliver a platform that multiple business units can adopt without coordination overhead
- Respond and improve metrics, traces, logs, IBM Instana, and incident workflows to cut mean-time-to-recovery and convert every incident into systematic improvement
- Secure and delegate IAM, secrets rotation, and posture management to enforce least-privilege access across humans and automation without slowing delivery
How to prepare for this exam
- Review the official exam guide to understand every objective and domain weight before you begin studying
- Work through the relevant IBM Training learning path — ibm certified advanced architect v2 plus ibm professional cloud sre v2 F1002700 — to cover vendor-authored material end-to-end
- Get hands-on inside IBM TechZone or a comparable sandbox so you can practice the console tasks, CLI commands, and APIs the exam expects
- Tackle a real-world project at your workplace, a volunteer role, or an open-source repository where the technology under test is actually in use
- Drill one exam objective at a time, starting with the highest-weighted domain and only moving on once you can teach it to someone else
- Study by objective in PowerKram learn mode, where every explanation links back to authoritative IBM documentation
- Switch to PowerKram exam mode to rehearse under timed conditions and confirm you consistently score above the pass mark
Career paths and salary outlook
Architect-plus-SRE generalists who own a platform end-to-end are rare enough to command premium pay:
- Platform Architect & SRE Lead — $160,000–$220,000 per year, owning platform design and operation end-to-end (Glassdoor salary data)
- Principal SRE — $155,000–$215,000 per year, setting reliability strategy across multi-cloud platforms (Indeed salary data)
- Advanced Cloud Architect — $165,000–$225,000 per year, designing enterprise-grade cloud platforms (Glassdoor salary data)
Official resources
Work through the official IBM Training learning path for this certification, which bundles videos, labs, and skill tasks aligned to every objective. The official exam page lists the full objective breakdown, prerequisite knowledge, and scheduling details.