20 Workstations for a University Lab: What It Takes to Do It on Budget

Updated 2026-08-19: the per-unit price in this scenario has been corrected twice. It originally carried a stale figure of ₦1.18 million per machine, far below what the components actually cost. The correction published on 2026-08-12 replaced it with ₦3,471,503, but that figure was itself computed from an internal pricing document carrying a doubled margin, making it roughly 11% too high. The per-unit figure below is ₦3,133,684, derived directly from the pricing engine our configurator quotes from. The change matters for the conclusion, not just the arithmetic: a ₦28 million allocation covers roughly seven machines at this specification, not the six to seven the previous version stated. If you took figures from this page for a procurement submission before 19 August 2026, please re-check them against a current quote.

The Scenario

Consider the Department of Computer Science at a federal university in Enugu, with a problem many Nigerian public institutions share: a legitimate need for modern computing infrastructure and a procurement budget approved two years earlier at prices that no longer reflect reality. The department head has been allocated funding for 20 student workstations as part of a lab upgrade initiative — say ₦28 million for hardware, installation, and a three-year maintenance agreement.

Approach the standard government-approved IT vendors with that and the quotes come back well above the allocation, for branded machines that are not significantly better than what the lab already has.

The Challenge

The lab's use case is specific: undergraduate computer science education. The machines need to run:

• Programming environments: Visual Studio Code, Python, Java IDEs, GCC toolchains
• Database tools: MySQL Workbench, PostgreSQL, MongoDB
• Simulation and data science: MATLAB, Jupyter notebooks, R Studio
• Virtualisation: VirtualBox for OS and networking coursework
• Occasional: Blender (for a digital art elective), Unity (for a game development course)

The machines will be used by students who are not always careful with hardware. They will run in a room air-conditioned by a single 3HP split unit of uncertain reliability. They need to be maintainable by the department's sole IT technician, who is competent but not a specialist. And they need to last at least five years under daily institutional use — something branded consumer hardware rarely achieves.

There is usually one more requirement, born of experience: a previous lab procurement where machines arrived that did not match the quoted specs. The department wants a vendor who will provide a detailed spec sheet before order and allow inspection on delivery.

The Assessment

A thorough look at the lab space surfaces the binding constraint. A room with one power circuit for the entire 20-station lab is a significant limitation — running 20 PCs simultaneously on a single circuit means designing for low power draw without sacrificing performance. Existing air conditioning that is barely adequate for the space is the other issue; a second unit requisitioned through a separate budget line, before installation, is the right sequence.

What the academic workload calls for: strong multicore CPU performance for compilation and virtualisation, adequate GPU for the occasional 3D coursework, and prioritised storage performance for database and data science work. A documented build specification, delivery manifest, and a three-year parts warranty on every machine are what the branded vendors typically do not offer at comparable prices.

The Build

20 Academic Workstations — ₦3,133,684 each, priced against Sephora's current catalogue:

• CPU: AMD Ryzen 7 7700 (₦703,517) — 8 cores, 65W TDP (important for the power circuit constraint), strong single-core for compilation
• Motherboard: B650 mATX DDR5 (₦411,145) — AM5, required for the Ryzen 7 7700
• RAM: 32GB DDR5-5600 (₦779,992) — sufficient for VirtualBox sessions and MATLAB without paging
• GPU: GeForce RTX 3050 8GB (₦489,038) — handles Blender and Unity coursework; supports dual monitors
• Storage: 1TB PCIe 4.0 NVMe SSD (₦215,001) — OS and software
• Cooling: Air Cooler · Basic (₦84,998) — silent, reliable, no liquid maintenance overhead
• PSU: 550W 80+ Bronze (₦149,999) — modular, ample headroom for a 65W CPU and 130W GPU
• Case: mATX Tower Case (₦299,994) — tool-free access panels, dust filters, easy maintenance

Per-machine UPS is priced separately from the workstation build above, as part of the lab's shared infrastructure budget below — sized for the combined CPU-and-GPU load rather than the bare minimum, with twenty units on a staggered power load to stay within the lab's single circuit.

What this means for the ₦28 million allocation: twenty workstations at this specification run to roughly sixty-three million naira in hardware alone — about two and a quarter times the entire budget, before a single UPS, monitor, or network cable is bought. The ₦4.4 million originally earmarked for a managed network switch, 20 24" 1080p monitors, structured cabling, per-machine UPS units, a shared 40TB NAS, and a three-year maintenance agreement doesn't close that gap; it barely dents it. Keep that ₦4.4 million reserve intact and the remaining budget buys roughly seven workstations at this specification — not twenty (that ₦4.4 million figure hasn't itself been re-verified against current prices in this pass; if it has drifted the way the workstation spec had, the real number of affordable units is lower still).

Dropping the discrete GPU — the single most expensive line item after the CPU — saves real money per unit, but nowhere near enough on its own to close a gap this size across twenty units. A full itemised spec sheet with component serial numbers before order confirmation is still what makes inspection on delivery possible, whatever unit count and spec the department settles on — the technician should be able to pull machines at random and check them against the sheet.

What Changes

Compilation times for student Java and C++ projects are the best proxy for general lab responsiveness, and they respond to single-core speed and core count together — which is why a current 8-core chip transforms the experience over the machines it replaces. MATLAB simulations scale similarly. VirtualBox sessions that cause older machines to swap and stall are stable with 32GB of RAM headroom, because the problem was never CPU — it was memory.

The maintenance change comes from the case, not the components. Magnetic dust filters and a positive-pressure airflow design keep dust out of a lab that would otherwise accumulate it, which means less reactive maintenance for a technician who is not a specialist and has twenty machines to look after.

Build Specification Summary

Procurement Realities: RFQ, Lead Time, and Warranty for Institutional Orders

An order this size — 20 machines, a shared NAS, monitors, and a multi-year maintenance agreement — runs through a different process than a single workstation purchase, and it's worth knowing what that process actually looks like before a department commits a budget line to it. Sephora Systems handles deployments from 3 to 300+ systems, with typical lead times of 7-21 days depending on volume and configuration — a department planning around a semester start date should request a quote with enough runway to account for the upper end of that range, particularly for an order that also includes networking hardware and a NAS with its own sourcing lead time.

Every build carries a 1-year local warranty covering manufacturing defects and component failures, handled in-country rather than requiring anything to ship abroad for repair — a meaningful difference for an institution whose sole IT technician doesn't have the bandwidth to manage an international RMA process on top of everything else on their plate. Institutional and multi-year deployments can extend this with a formal maintenance agreement and SLA terms for guaranteed response times, which is the mechanism behind the three-year maintenance line in the budget above.

For government-funded and tender-based procurement specifically, the process runs through Sephora's enterprise inquiry rather than a retail checkout: the form captures organisation, unit count, use case, budget band, and timeline in one submission, and supports purchase-order and tender-based procurement rather than assuming a direct card payment. A documented, itemised spec sheet — the same one that makes delivery inspection possible, as described above — is a standard part of a quote through this path, not a special request.

Does This Scale Beyond 20 Seats?

The same standardisation logic that makes 20 workstations manageable for one technician applies whether the real number is 10 or 60. Our fleet-standardisation guide covers why ordering a small number of known configurations — rather than a spec that drifts machine to machine — is the single highest-leverage decision for any multi-seat institutional order, lab or otherwise. And the sourcing reality behind a 20-unit order is the same one behind any bulk order in Nigeria: our bulk procurement guide covers how a multi-unit quote actually gets built and delivered, and why a single dated, itemised quote is the honest way to price a fleet against a market that mostly doesn't publish per-component retail prices.

The budget-cycle mismatch this scenario opens with — money allocated against prices that were current years earlier — isn't unique to universities; it's a structural feature of any institutional or public-sector procurement with a multi-year approval cycle. Our total-cost-of-ownership breakdown is worth reading alongside this scenario for exactly that reason: a budget planned once and executed years later needs to account for both sourcing price movement and the multi-year running cost of whatever gets bought, not just the sticker price at approval time. Requesting a current, dated quote before a budget cycle locks in is the single cheapest way to avoid discovering the gap only after the money is already committed.

Key Takeaway

University and institutional procurement in Nigeria is constrained by budget cycles and approval processes that often mean the money available doesn't match current hardware prices — and this scenario is a real example of that gap, not a hypothetical one. At today's prices, twenty workstations at the assessed specification cost roughly two and a half times the entire allocation, in hardware alone. That gap isn't closed by shopping harder or negotiating a discount; it's closed by one of three decisions: buy fewer machines now and phase the rest against a future budget cycle, reduce the specification and accept the trade-off it means for GPU-dependent coursework, or return to the funding body with a current, dated quote before the allocation is locked in. Whichever the department chooses, the sequence matters: get a real quote before the budget cycle commits a number, not after. A vendor relationship built on documented specs and accountable delivery only helps once the budget itself reflects reality.

Running an educational institution or research facility that needs computing infrastructure? Talk to our team, or submit an enterprise inquiry directly if you already have a unit count and timeline in mind — we work with universities, schools, and research labs on spec-to-installation projects.