What this assignment does
The report evaluates three architecture options for a UK grocery retailer's analytics platform, proposes one, and sets out the cost-control practices needed to keep it viable.
Its most useful section refuses the standard claim that cloud is cheaper, and identifies the usage pattern that decides it.
How it is built
Literature, comparative evaluation of three options, proposed architecture, critical discussion, recommendations. The options are scored against the retailer's actual workload rather than against generic criteria.
The business is concrete throughout — a Christmas trading peak, spiky demand, steady always-on reporting — which is what allows the evaluation to reach a decision.
Annotated extracts
Three passages, each with a note on why it earns marks and what a weaker answer does in the same place — enough to teach the move, not enough to stand in for the assignment.
The cloud's pay-as-you-go model is often presented as cheaper, but this depends heavily on usage patterns. Elastic compute is economical for spiky workloads, such as the Christmas trading peak, but steady always-on workloads can cost more in the cloud than on owned hardware. Uncontrolled queries and idle clusters are common causes of budget overruns.
Why this marks well
The report takes the central marketing claim of its own subject and makes it conditional. Cloud is cheaper for spiky workloads; for steady ones it can be more expensive than owning the hardware.
The condition is tied to this business — a grocery retailer has a genuine Christmas peak, which is exactly the pattern that makes elasticity pay. The analysis is therefore a finding about FreshCart, not a general observation about cloud computing.
Naming the two common overrun causes — uncontrolled queries and idle clusters — is what makes the following FinOps recommendations concrete rather than a list of best practices.
What a weaker answer does here
A weaker answer states that cloud reduces capital expenditure and offers elastic scaling. Both are true, neither is conditional, and the report cannot then explain why any organisation still runs its own hardware.
Three options were evaluated against the retailer's workload: batch processing of nightly sales data, near-real-time stock monitoring, and ad-hoc analyst queries.
Why this marks well
The evaluation criteria are three named workloads rather than abstract qualities like scalability or performance. That is what makes the comparison decidable — each option can be strong on one workload and weak on another.
It also forestalls the commonest failure in architecture coursework, which is choosing the most capable platform. The best option here is the one that fits three specific demand profiles, not the one with the longest feature list.
Because the workloads are stated, a marker can disagree with the conclusion on the evidence, which is the sign of a genuinely assessable argument.
What a weaker answer does here
A weaker answer compares Hadoop, Spark and a managed cloud service on general characteristics in a table. The comparison is accurate, applies to any organisation, and therefore cannot select for this one.
FreshCart should therefore adopt FinOps practices: tagging all resources by team and use case, setting budgets and alerts, using auto-termination on clusters, and purchasing reserved capacity for predictable baseline load.
Why this marks well
The recommendations are operational and specific enough to implement on Monday. Tag by team and use case; set budgets and alerts; auto-terminate clusters; reserve capacity for the baseline.
The last one closes the loop with the earlier analysis. Reserved capacity for predictable baseline load is precisely the answer to the finding that steady workloads cost more in the cloud — the report has used its own critique to shape the design.
Cost control is also presented as part of the architecture rather than as a management afterthought, which reflects how cloud platforms actually fail in practice.
What a weaker answer does here
A weaker answer recommends monitoring costs carefully. Nobody disagrees, nobody can act on it, and it does not respond to anything the report found.
What makes this a strong answer
It makes the central claim of its own field conditional, and identifies the condition from the client's actual demand profile.
Options are evaluated against three named workloads, so the comparison can reach a decision rather than a table.
The cost critique feeds directly into the proposed design instead of sitting in a closing caveat.
Request the full document
Request the full document
The extracts above are the teaching material and they stay on this page. The complete 2,050-word document is sent person to person, by a human, after a short conversation about what you are working on — so that we can point you at the right parts of it rather than hand over a file and wish you luck.
Both links below already name this sample (DSC-04), so you do not have to describe it.
WhatsApp · +44 7537 126 976 · Email · mia@globalprojectshelp.com
Mon–Sat, 9–22 GMT. No form, no mailing list, no download gate.
Full documents are never published, hosted or linked on this site.
Using this sample honestly
This is a worked example. It sits in the same place as a worked example in a textbook or a past paper with an examiner’s report attached: read it, take the move, then write your own paragraph without it open in another window.
Submitting any part of this document as your own work would be plagiarism — the same offence as copying from a journal article, and detected the same way. If you draw on it, cite it. If a deadline has got to the point where submitting someone else’s paragraph feels like an option, say so when you message us; that is a problem with a legitimate answer and we would rather work on it than watch it happen.
Our full position, including how to declare mentoring support to your institution, is on the academic integrity policy page.
Where to go next
- More MSc Data Science samples The full set of annotated model answers for this course.
- All samples Every annotated model answer in the library, with discipline filters.
- How requesting works What you receive, what you do not, and the conditions attached.
- Academic integrity policy What we do, what we will not do, and how to declare mentoring support.
Model answer written by a Global Projects Help mentor · © Global Projects Help
Published 8 October 2026 · Questions? mia@globalprojectshelp.com