Summary

  • A vendor case study identifies Prasanna Premachandra as the ICT project manager behind a school wireless project designed around consistent access, movement between classrooms and operational simplicity.
  • The evidence supports a historical case about infrastructure choices and institutional capacity; it does not establish the school’s present architecture, Premachandra’s current control of the network or independently audited performance.

A lesson begins before the application loads

The cost of a school network failure is rarely confined to the network. A student reconnects. A teacher waits. A class loses the opening minutes in which attention is easiest to organise. Support staff receive another report whose technical symptom is simple but whose institutional effect is distributed across rooms, schedules and people.

A Fortinet customer case study about Melbourne Girls Grammar School makes that translation unusually visible. The document identifies Prasanna Premachandra as the school’s ICT project manager. It says the existing wireless network did not provide consistent access: more than 700 of 1,200 students lacked wireless coverage, and students who moved between classrooms could lose their connection and have to log in again. Because the account was published by the selected vendor, its figures and performance descriptions should be read as attributed project evidence, not as an independent audit.

Even with that limitation, it records the problem in operational terms that matter beyond one product.

The first constraint was coverage, but coverage alone was not the service. The school’s notebook programme made Internet access part of ordinary classroom work. Video and other bandwidth-intensive applications increased the need for throughput. Movement across the campus meant that a connection had to persist while a learner changed location. The network therefore had to coordinate three things at once: physical reach, shared capacity and continuity during movement.

Premachandra’s quoted requirement in the case study was for uninterrupted wireless access across 60 classrooms and for all teachers and students. That is a useful management formulation because it begins with the population and the activity, not with a device count. It defines success as access that survives the rhythm of the institution.

Fairness is an infrastructure property

Schools expose a problem that commercial network reporting can obscure. Average performance can look acceptable while access remains uneven. A well-served room and a poorly served room may produce a reasonable campus-wide average, but the students in the latter room still experience a different curriculum. If an older laptop consumes disproportionate airtime, users with newer equipment may also feel the effect even though total capacity has not changed.

The Fortinet case study describes the project as addressing this through airtime fairness and a single-channel wireless design. Those are vendor-characterised mechanisms, and the available public material does not provide independent packet captures or before-and-after service-level data. The institutional principle is nevertheless broader than the product claim: shared infrastructure needs an explicit rule for how scarce capacity is distributed.

That rule affects more than speed. When access is treated as a first-come or strongest-device contest, the network can reproduce differences in equipment, location and timing. When the design attempts to allocate airtime fairly, it acknowledges that the service is collective. The decision is therefore partly technical and partly about who bears congestion.

Premachandra’s role matters here because infrastructure leadership is the work of converting an institutional goal into operating requirements. “Wireless everywhere” is too vague to procure or test. Coverage across named spaces, concurrent access, movement without repeated login, support for existing devices and an architecture the internal team can manage are decisions. Once expressed, they can be piloted and challenged.

The pilot as a control mechanism

According to the case study, the school worked through Wavelink Communications and an integration partner on a three-month trial in the science and technology building before selecting the wider deployment. The document then describes a main controller, a backup controller and 60 access points supporting about 1,200 student laptops.

A pilot does not remove vendor risk, but it changes the evidence available to the buyer. Marketing claims become observable behaviour in a real building. Support staff can see whether devices reconnect, whether dense areas remain usable and whether ordinary changes require specialist intervention. Teachers can report whether a technical improvement actually reduces disruption. The institution also learns which measurements it failed to define before the trial.

The most important feature of a pilot is reversibility. A school can test assumptions in a limited environment before making campus-wide coverage dependent on them. That matters because wireless systems are shaped by walls, room density, client behaviour, interference and the movement patterns of real users. A design that works in a diagram may fail at the beginning of a lesson when hundreds of devices become active together.

The public evidence does not disclose the school’s procurement scoring, contract terms or alternative bids. It cannot show whether the selected design was the only viable option. It does show that the decision moved through a bounded trial rather than directly from specification to full deployment. For infrastructure leaders, that sequence is itself a control: test the mechanism where failure is containable, then expand only after the operating team understands what it is accepting.

Manageability determines the life of the system

Networks are often purchased at their moment of maximum attention and operated during years of relative invisibility. The staff who add an access point, investigate a coverage hole or replace failed hardware inherit the consequences of the original architecture. A system that performs well but requires rare expertise for routine adjustment can become progressively harder to sustain.

Premachandra emphasised ease of installation and management in the vendor case study. The document says the single-channel design reduced the need for repeated channel planning when access points were added or moved. It also reports that seven legacy access points in the library were replaced by two units under the new design. These are vendor-reported outcomes, not independently verified cost savings. They nevertheless illustrate the operating question behind capital expenditure: how much ongoing coordination does the design create?

Fewer components do not automatically mean lower risk. Concentration can increase the impact of a failure, and proprietary architecture can deepen switching costs. The case study’s reference to a main and backup controller suggests that continuity was considered, but the public document does not reveal failover tests, maintenance procedures or dependency on vendor support. A responsible reading holds both sides together. Simplification can reduce daily operating burden, while centralisation can make resilience and exit planning more important.

This is where leadership extends beyond choosing a product. The durable work includes documenting coverage intent, preserving configuration knowledge, testing recovery, training staff and ensuring that a future team can distinguish a radio problem from authentication, upstream or device failure. When those practices are absent, a manageable design can become unmanageable through staff turnover and undocumented change.

What the registry record adds—and what it does not

Public routing records provide a separate link between Premachandra and the school’s Internet infrastructure. Data for AS138201 identifies Melbourne Girls Grammar School and reproduces an APNIC person record naming Prasanna Premachandra under handle PP516-AP. The person record was last modified in 2011. It is therefore evidence of a historical association in the Internet-numbering system, not evidence that he currently operates the autonomous system or controls the school’s routing.

That narrow evidence still matters. It places the same exact name found in the project case study inside a formal network-resource context. It also illustrates why registry data should be interpreted carefully. A person contact can remain visible after roles, organisations or procedures change. The record can support identity continuity while remaining silent about present authority.

A public professional profile under Premachandra’s name describes more than 30 years in IT and a current association with Luther College in the Greater Melbourne area. That profile is self-published career evidence, and it should not be used to infer the present network design of either institution. Read together, the sources show a career that continued beyond the historical project. They do not turn a dated deployment into a current case study.

Infrastructure leadership is measured in recovered attention

The strongest lesson in the public record is not that one wireless architecture should be copied. It is that infrastructure quality can be measured by the attention it returns to the institution. A stable connection returns minutes to teachers. Predictable mobility removes a repeated decision from students. Manageable access points reduce the number of specialist interventions required from support staff. A pilot returns evidence to the buyer before a larger commitment becomes difficult to reverse.

Those gains are easy to miss because they appear as absences: no cable carried between desks, no login repeated at the classroom door, no lesson paused while a device reconnects. Conventional technology budgets record equipment and licences more readily than recovered attention. Yet the latter is closer to the purpose of the investment.

Premachandra’s historical project offers a leadership case precisely because the documented requirements connect engineering choices to institutional time. The available evidence is bounded and vendor-mediated. It does not justify a general performance verdict. It does support a durable conclusion: when connectivity becomes part of learning, network design is no longer a back-office convenience. It is a decision about access, continuity and how much human attention the institution is willing to spend compensating for avoidable friction.

Sources