AWST · two hours behind, and it matters

SQL Server DBA services in Perth

Perth runs two hours behind the eastern states in winter and three in summer, which makes an eastern provider’s "after hours" your working afternoon. Remote SQL Server DBA cover with a response SLA measured in elapsed time, not business hours — from $1,500 a month.

Free 20-point health check Call 1800 431 416

Western Australia is on AWST, UTC+8, with no daylight saving. That puts Perth two hours behind Sydney and Melbourne for roughly half the year and three hours behind for the other half — the largest standing time difference inside Australia, and the one that does the most damage to a support arrangement designed around eastern business hours.

The consequences are specific. Your 8am is 10am or 11am in the east, so a problem raised first thing lands mid-morning for an eastern team. Your 3pm is their 5pm or 6pm, so anything raised in your working afternoon arrives as they are leaving. And your overnight change window sits squarely inside the eastern small hours, which is precisely where a provider running "business hours plus best efforts" has the least to offer.

The answer is not pretending the offset does not exist. It is a response time measured in elapsed hours under an SLA, rostered to be answered whenever the clock runs out — which is what a 24/7 plan is actually for.

What the offset actually costs you

Why "we support Australia" is not the same as supporting Perth

Plenty of eastern providers list WA on the website. The question to ask is what happens at 3pm your time, and what happens during your change window.

AWST is UTC+8 with no daylight saving — two hours behind AEST, three behind AEDT
A response SLA expressed in elapsed hours is unaffected by the offset; one expressed as "business hours" quietly becomes next-day cover for a Perth client raising anything after about 2pm
Your overnight change window falls inside the eastern small hours, so it needs a rostered on-call engineer rather than goodwill
Onsys plans guarantee a response in one to two hours depending on tier, at any hour, measured from when the alert or ticket is raised
There is no major public cloud region in Western Australia — Perth workloads in Azure or AWS are almost always served from Sydney or Melbourne, which is a design constraint worth being explicit about
Where latency to an eastern region is a real problem for your application, that is an architecture conversation and we will have it honestly rather than sell you a plan that cannot fix it

What we get called about in WA

Resources, remote sites and systems that never stop

Western Australian estates are shaped by continuous operations and long distances, and the database problems follow from that.

No maintenance window at all

Continuous operations in resources, utilities and logistics, where there is no overnight quiet period because there is no overnight. Patching becomes a high-availability design problem rather than a scheduling one.

Databases at remote sites

Instances at mine sites, ports and processing facilities, on connectivity that constrains the recovery strategy far more than anyone assumed when the backup job was written.

Latency to an eastern region

With no cloud region in WA, a Perth application talking to a database in Sydney pays for every round trip. Sometimes that is fine; sometimes it is the whole problem, and the fix is in the application’s query pattern.

Operational technology databases

Historians, SCADA back-ends and plant systems on SQL Server, often owned by engineering rather than IT, and often outside whatever the corporate patching regime covers.

A DR site three hours away by clock

Replication to an eastern site, where the two ends disagree about what time it is for half the year and the failover has never been rehearsed against the real workload.

WA government security expectations

Western Australian public sector entities work to the state cyber security policy, which, like the ACSC Essential Eight, expects vendor-supported and patched software. An ageing SQL Server version is the control that fails quietly.

Being straight about this

We are not based in Perth

Onsys is a Melbourne practice. We support Western Australian organisations remotely under a 24/7 SLA measured in elapsed time, which is the only arrangement that survives a three-hour offset.

Our office is at Level 1, 530 Little Collins Street, Melbourne VIC 3000, and our DBAs are in Australia
Database support is remote, which is how specialist database work is delivered everywhere — the alternative is an in-house DBA, not a local one
On-site attendance is available in the Melbourne metropolitan area as part of an engagement; elsewhere it is arranged and quoted case by case
What you get instead of proximity is a guaranteed response time, a named consultant who knows your estate, and published prices
If physical attendance matters more to you than any of that, we will say so rather than talk you out of it

Start with one Perth instance

The free 20-point health check is read-only and needs no access to your environment — you run the scripts in your own time, on your own clock, and we send the written findings within 3 business days.

Book the free health check

Published 9 October 2026

FAQ

Common questions

No. Onsys is a Melbourne practice and our DBAs are in Australia. Western Australian clients are supported remotely under a 24/7 response SLA measured in elapsed time, so the two or three hour offset does not extend the time it takes to get a DBA on your problem. On-site attendance outside the Melbourne metropolitan area is arranged and quoted case by case.

It does not. The SLA is one to two hours depending on tier, measured from when the alert fires or the ticket is raised, 24 hours a day including weekends and public holidays. That is a rostered on-call arrangement rather than business-hours cover, which is precisely why it works across a three-hour offset. If a provider quotes you a response time in business hours, ask which business hours they mean.

No — that is the normal case for WA clients and it is planned for rather than tolerated. Changes are scheduled in AWST, carry a tested rollback, and are staffed by the on-call roster. Where you have no maintenance window at all, which is common in continuous operations, patching becomes a high-availability design question and we will address it that way.

Not usually, but it has to be a decision rather than an accident. A Perth application talking to a database in Sydney pays a latency cost on every round trip, and whether that matters depends entirely on how chatty the application is. For some workloads it is unnoticeable; for others it is the whole problem, and the fix is in the query pattern rather than the hosting. We will tell you which one you have before you commit to a migration.

Plan A is $1,500 per month excluding GST for up to 10 SQL Server instances — $150 per instance — with a guaranteed two-hour response at any hour and 10 professional service hours a month. Plan B is $3,000 with a one-hour SLA and Plan C is $7,500 for larger mixed-platform estates. There is no loading for being in Western Australia.