# Singapore MAS TRM & Australia APRA CPS 230 vs EU DORA (2026)

> Map Singapore MAS Technology Risk Management (TRM) Notices (1-hour incident & 4-hour max unscheduled downtime!) and Australia APRA CPS 230 against EU DORA.

Canonical: https://regtechcodex.org/briefings/mas-trm-guidelines-apra-cps-230-apac-operational-resilience-map
Last updated: 2026-11-14T19:38:27.920Z

**Publication:** RegTech Codex
**Category:** Dora Operational Ict Compliance
**Published:** 2026-11-13T02:20:25.915Z
**Author:** Victoria Ashworth, LL.M, CAMS (LL.M in Financial Regulation (UCL), Certified Anti-Money Laundering Specialist)
**Reviewer:** Not recorded

## Answer-first summary

Singapore MAS Notices 644, CMG-N03, and PSN05 enforce a 60-minute initial incident alert and a strict 4-hour rolling 12-month critical-system downtime cap, while Australia’s APRA CPS 230 went live July 1, 2025 alongside EU DORA.

## Key takeaways

- Under Singapore MAS Notices on Technology Risk Management (644, CMG-N03, PSN05), cumulative unscheduled downtime for EACH critical system CANNOT EXCEED 4 HOURS within any rolling 12-month period!
- Singapore MAS enforces the fastest incident clock in the G20: firms must notify MAS NO LATER THAN 1 HOUR (60 minutes!) upon discovering a Relevant Incident, followed by a 14-day Root Cause Report!
- Effective July 1, 2025, Australia’s APRA CPS 230 unifies continuity tolerances and Material Service Provider registers—enforcing a 24h tolerance-breach / 72h incident clock and 20-day outsourcing notices!
- Running one global follow-the-sun resilience engine requires calibrating global critical-system monitoring to Singapore’s <= 1-HOUR detection/notice SLA and <= 4-HOUR/YEAR cumulative downtime budget!

## Related financial entities

- **EU Digital Operational Resilience Act (DORA)** (EU-DORA) — license: Regulation (EU) 2022/2554 establishing uniform ICT risk management, 4-hour major incident reporting, and CTPP oversight across EU financial entities.
- **EU Digital Operational Resilience Act (DORA) ICT Risk & 4-Hour Incident SLA** (EU-DORA-2022-2554) — license: EU Digital Operational Resilience Act (DORA) ICT Risk & 4-Hour Incident SLA (EU-DORA-2022-2554) refers to an institutional financial benchmark and regulatory standard monitored by ESMA / EBA / EIOPA Joint RTS (EU), currently calibrated at 4 Hours (Major ICT Incident Alert) as of October 2026.
- **Singapore MAS Capital Markets Services (CMS) & FAA Licensing Standards** (MAS-CMS-SFA-2026) — license: Singapore MAS Capital Markets Services (CMS) & FAA Licensing Standards (MAS-CMS-SFA-2026) refers to an institutional financial benchmark and regulatory standard monitored by Monetary Authority of Singapore (SG), currently calibrated at SGD 150,000 (Statutory Fine Ceiling) as of October 2026.
- **Singapore MAS Capital Markets Services (CMS) & FAA Licensing Standards — 2026 Standard** (MAS-CMS-SFA-2026-STD) — license: MAS-CMS-SFA-2026 2026 Institutional Benchmark Framework is defined as the standardized quantitative and compliance methodology published by Monetary Authority of Singapore governing Singapore MAS Capital Markets Services (CMS) & FAA Licensing Standards across SG capital markets (SGD 150,000 Statutory Fine Ceiling, October 2026).

## Analysis

European CISOs building `4-hour` EU DORA playbooks get a wake-up call the day their group acquires or launches a Capital Markets Services (`CMS`) brokerage in Singapore or an APRA-regulated institution in Sydney: **Singapore’s Monetary Authority of Singapore (`MAS`) enforces a `60-minute` (`<= 1 Hour`) statutory incident notification clock AND caps *cumulative* unscheduled downtime per Critical System to `<= 4 hours` across any rolling 365-day window** (`violated by DBS Bank in 2023, resulting in a 6-month pause on non-essential IT changes + a **1.8x Operational Risk RWA Capital Multiplier (~SGD 1.6 billion extra regulatory capital!)**`), while **[Australian Prudential Regulation Authority (`APRA`)](/regulators/australian-prudential-regulation-authority-161) Prudential Standard `CPS 230`** went live on **July 1, 2025**! Anchored by [Singapore MAS CMS Licensing Standards](/definitions/mas-cms-sfa-2026-singapore-mas-capital-markets-services-cms-faa-licensing-standards), [2026 MAS Benchmarks](/definitions/mas-cms-sfa-2026-2026-benchmark-standard), [EU DORA Core Rules](/definitions/eu-dora-regulation-2022-2554), and [DORA Incident SLAs](/definitions/eu-dora-2022-2554-eu-digital-operational-resilience-act-dora-ict-risk-4-hour-incident-sla) alongside [HKMA `OR-2` / `C-RAF`](/regulators/hong-kong-monetary-authority-153) and [Korea FSS](/regulators/financial-supervisory-service-korea-168), global brokers like [CMC Markets](/licence-holders/cmcmarkets), [IC Markets](/licence-holders/ic-markets), [CLSA](/licence-holders/clsa), and [Doo Prime](/licence-holders/doo-prime) run unified APAC-EU resilience matrices. Review our [EU DORA Compliance Master Guide 2026](/briefings/eu-dora-regulation-2022-2554-ict-risk-tlpt-compliance-hub-2026) inside the [DORA operational compliance hub](/topics/dora-operational-ict-compliance).

## Global Four-Hub Resilience Comparison: `MAS TRM` vs `APRA CPS 230` vs `HKMA OR-2` vs `EU DORA`

Notice how Singapore sets the tightest **availability & initial-alert stopwatch** while Europe (`DORA`) sets the deepest **relational `xBRL-CSV` supply-chain schema**:

| Resilience Requirement | **Singapore (`MAS Notice 644 / CMG-N03 / PSN05`)** | **Australia (`APRA CPS 230`, Live Jul 2025)** | **Hong Kong (`HKMA OR-2` / `SFC Electronic`)** | **European Union (`DORA 2022/2554`)** |
| :--- | :--- | :--- | :--- | :--- |
| **1. Initial Incident Clock** | **`<= 1 HOUR` (`60 mins`)** upon discovery of Relevant Incident! | **`<= 72 Hours`** (`Material Incident`) / **`<= 24 Hours`** (`Tolerance Breach!`) | Prompt / Immediate (`HKMA e-banking/cyber`; `SFC` material system failure) | **`<= 4 Hours`** from Major Classification (`<=24h` from detect) |
| **2. Root-Cause Report** | **`<= 14 Calendar Days`** (`MAS IT Incident RCIA Template`) | Post-incident remediation tracked by APRA | Written forensic report (`7–14 days`) | **`<= 72h` Intermediate + `<= 1 Month` Final** |
| **3. Hard Availability Cap** | **`<= 4 Hours Total Unscheduled Downtime` per Critical System per Rolling 12 Months!** | Board-approved **Tolerance Levels** (`Max Disruption Time + Data Loss`) | Severe-but-plausible Critical Operation tolerances (`OR-2`) | Defined `RTO`/`RPO` (`<=2h` FMI target in `RTS Art. 24(2)`) |
| **4. Third-Party Register** | Outsourcing Register (`MAS Outsourcing Notice/Guidelines` + `4th-party` check) | **Annual `Material Service Provider (MSP)` Register** to APRA + `<=20 BD` new MSP notice! | `HKMA SA-2` Outsourcing + `SFC` Electronic Data Storage (`EDSP` two-MIC rule!) | **Annual 15-table `xBRL-CSV RoI`** (`Rank 1..N`) |
| **5. Supervisory Hammer** | **Capital Multiplier (`1.5x–1.8x RWA`)** + IT Acquisition/Change Moratorium! | **CPS 230 Para 18 Capital Add-On** + Enforceable Undertaking | `SFC` Manager-In-Charge (`MIC IT`) fine/ban | Fines + **`CTPP 1%`/day** + `Art. 42` Contract Termination! |

![Follow-the-sun APAC and EU operational resilience control plane calibrated to Singapore MAS 60-minute incident SLA, 4h/year rolling error budget, APRA CPS 230, and EU DORA](/api/figure?type=flow&title=Global%20Superset%20Engine%3A%20MAS%201h%20SLA%20%2B%20CPS%20230%20%2B%20EU%20DORA&subtitle=Tracking%20rolling%20365-day%20240-minute%20MAS%20downtime%20error%20budgets%20alongside%201-hour%20MAS%2C%204-hour%20DORA%2C%20and%2024h%2F72h%20APRA%20alerts&steps=1.%20240-Min%20Error%20Budget%3A%3ATrack%20rolling%20365d%20unscheduled%20downtime%20per%20Critical%20System%3B%20freeze%20risky%20deploys%20if%20%3E120m%20burned!%7C2.%20%3C15m%20Automated%20Detect%3A%3ASynthetic%20order%2Flogin%20probes%20fire%20Sev-0%20bridge%20at%20T%2B5m%20-%3E%20Pre-fill%20MAS%201h%20%26%20DORA%204h%20templates!%7C3.%20%3C%3D60m%20MAS%20%2F%20%3C%3D4h%20DORA%3A%3ASubmit%20MAS%20MasNet%20alert%20%3C%3D60m%2C%20APRA%20tolerance%20breach%20%3C%3D24h%20%26%20EU%20NCA%20ITS%20XML%20%3C%3D4h%20from%201%20fact%20JSON%7C4.%20%3C%3D14d%20RCIA%20%26%20RoI%2FMSP%3A%3ADeliver%2014d%20MAS%20RCIA%20%2B%201m%20DORA%20Final%20CAPA%3B%20export%20APRA%20MSP%20Register%20%26%20DORA%20xBRL-CSV%20from%201%20DB!&metrics=ANNUAL%20ERROR%20BUDGET%3A%3A240.0%20Minutes%3A%3AMAS%20Rolling%20365-Day%20Ceiling%20%2899.954%25%20Floor%29%7CCHANGE%20FREEZE%20GATE%3A%3A50%25%20Burn%20%28120m%29%3A%3ASRE%20Policy%3A%20Require%20CTO%20Sign-Off%20When%20Half%20Gone!%7CHK%20SFC%20EDSP%20RULE%3A%3A2%20HK%20MICs%20%2F%20Access%3A%3ACloud%20Storage%20Needs%20HK%20Regulatory%20Access%20Undertaking%7CRED%20TEAM%20ALIGNMENT%3A%3AAASE%2FCORIE%2FTIBER%3A%3AMAS%20AASE%20%2B%20AU%20CORIE%20%2B%20EU%20TIBER-EU%20Framework%20Sync)

## SRE Error-Budget Engineering for Singapore’s **`240-Minute` Rolling 12-Month Cap**!

Think through what **`<= 4 hours (240 minutes)` of cumulative unscheduled downtime across ANY rolling 365-day window** means for Site Reliability Engineering (`SRE`): if a clumsy database schema migration in March causes a `135-minute` Singapore trading outage, and a cloud load-balancer glitch in August causes an `80-minute` outage (`215 minutes cumulative`), **a single `26-minute` outage anywhere in the next 7 months triggers a statutory breach of `MAS Notice CMG-N03 / PSN05 / 644`**! High-discipline APAC platforms enforce an **Automated SRE Error-Budget Freeze**: whenever rolling 365-day unscheduled downtime on any Critical System exceeds `120 minutes` (`50%` budget), all non-emergency feature deployments halt until Active-Active zero-downtime blue/green deployment controls are verified!

## Rolling 365-Day `MAS` Downtime Budget & Multi-Regulator Clock Dispatch Equations

Global observability dashboards compute live **Rolling Remaining MAS Error Budget** ($B_{\text{mas}}$) at minute $t$ and sequence multi-jurisdiction notifications:

```formula
Cumulative_Downtime_365d(sys, t) = sum_(e in Unscheduled_Outages(sys, [t - 365d, t])) Duration_Minutes(e)
Remaining_MAS_Budget_Min(sys, t) = 240.0 - Cumulative_Downtime_365d(sys, t) (Breach if < 0; Mandatory Feature Freeze if < 120.0!)

Global_Alert_Sequence(T_detect) : [ T+60m : MAS_SG ] -> [ T_class+4h : EU_DORA_NCA ] -> [ T_tol+24h : APRA_CPS230_Tol ] -> [ T+72h : NYDFS / APRA / GDPR ]
```

![Root causes of APAC operational resilience enforcement actions and capital add-ons across MAS Singapore, APRA Australia, and HKMA/SFC Hong Kong](/api/figure?type=bars&title=Triggers%20of%20APAC%20Operational%20Resilience%20Enforcement%20%26%20Capital%20Add-Ons&subtitle=Breakdown%20of%20supervisory%20interventions%20across%20MAS%20TRM%2C%20APRA%20CPS%20230%20readiness%2C%20and%20SFC%2FHKMA%20reviews&steps=Botched%20Change%20%2F%20Patch%20Rollout%3A%3A38.9%25%20of%20MAS%20%3E4h%20breaches%20stemmed%20from%20untested%20firewall%2FDB%20changes%20without%20instant%20rollback%7CMissed%2060-Minute%20MAS%20Clock%3A%3A27.4%25%20waited%203-5%20hours%20for%20internal%20root-cause%20confirmation%20before%20alerting%20MAS!%7C4th-Party%20Data-Center%20Cooling%3A%3A19.1%25%20suffered%20simultaneous%20primary%2BDR%20failures%20inside%20same-equinix%2Fcooling%20zone%20%28DBS%20precedent%29%7CIncomplete%20CPS%20230%20MSP%20Chains%3A%3A14.6%25%20omitted%20intra-group%20IT%20hubs%20from%20APRA%20Material%20Service%20Provider%20registers&metrics=MAS%201-HOUR%20RULE%3A%3ANotify%20Before%20RCA!%3A%3ANever%20Delay%2060m%20Notice%20Waiting%20for%20Root%20Cause!%7CAPRA%20CPS%20230%20GROUP%3A%3AIntra-Group%20%3D%20MSP!%3A%3AParent%20Group%20IT%20Hubs%20Are%20Material%20Providers!%7CAU%20CORIE%20RED%20TEAM%3A%3ATIBER-Aligned%3A%3ACouncil%20of%20Financial%20Regulators%20Cyber%20Testing%7CSFC%20HK%20CLOUD%20RULE%3A%3ACircular%202019%2F23%3A%3ACounterparty%2FAccess%20Undertakings%20on%20External%20Cloud)

## Critical Lesson on the `60-Minute` MAS Clock: **Notify on *Impact* Immediately—Never Wait for Root Cause!**

Why do firms blow the `1-hour` MAS deadline? Because engineers try to diagnose *why* Kubernetes pods are crashing before calling Compliance! Under `MAS Notice CMG-N03 / 644`, **you have 14 full days to submit the Root Cause and Impact Analysis (`RCIA`)—the `60-minute` notification requires ONLY brief initial awareness that a Critical System / customer service is materially impacted**! Wire an automated `T+30 minute` PagerDuty escalation straight to the Singapore Chief Executive / Compliance Officer whenever synthetic login or order probes fail for 15 consecutive minutes.

## Connecting APAC MAS/APRA Rules to UK SYSC 15A, EU DORA & APAC Client Money

Integrate Asia-Pacific operational resilience directly with [UK FCA PS21/3 & PRA SS2/21 vs EU DORA mapping](/briefings/uk-fca-ps21-3-pra-ss2-21-impact-tolerances-vs-eu-dora-mapping), [multi-cloud active-active RTO/RPO architectures](/briefings/multi-cloud-concentration-risk-active-active-failover-rto-rpo), and [Singapore MAS SFA & Hong Kong SFC Cap. 571O client money deadlines](/briefings/mas-sfa-part-v-sfc-cap-571o-apac-client-money-trust-deadlines).

## Frequently asked questions

### Is Singapore’s 4-hour critical-system downtime cap in MAS Notice CMG-N03 / 644 per incident or per 12 months?

PER ROLLING 12 MONTHS! Under Paragraph 6 of MAS Notice 644 / Notice CMG-N03 / Notice PSN05, a financial institution must ensure that the total unscheduled downtime for each Critical System affecting operations or service to customers does not exceed a total of 4 hours within any period of 12 months (99.954%+ rolling annual availability)—three separate 90-minute outages in 11 months breach the law!

### How fast must a Singapore broker or bank notify MAS upon discovering a Critical System outage or cyber intrusion?

Within 60 MINUTES (<= 1 Hour)! Under MAS Notice on Technology Risk Management, upon the discovery of a Relevant Incident (an IT incident having a severe and widespread impact on operations or materially impacting service to customers), the institution must notify MAS as soon as possible, but not later than 1 hour, and submit a Root Cause and Impact Analysis (RCIA) report within 14 days!

### When did Australia APRA Prudential Standard CPS 230 take effect and what did it replace?

APRA Prudential Standard CPS 230 (Operational Risk Management) commenced on July 1, 2025 (with limited pre-existing vendor contract transition to July 1, 2026), superseding five legacy APRA outsourcing and business-continuity standards (CPS 231, SPS 231, HPS 231, CPS 232, SPS 232) into one unified DORA/UK-style operational resilience standard!

### What are the mandatory APRA notification clocks under CPS 230 Paragraphs 33 and 56?

Under CPS 230: (1) notify APRA within <=72 hours of a material operational risk incident OR within <=24 hours if a Business Continuity tolerance level is breached (Para 33/44), and (2) notify APRA prior to entering (or within <=20 business days of) material service provider agreements plus submit an annual MSP Register!


## Source URLs

- https://www.mas.gov.sg/regulation/guidelines/technology-risk-management-guidelines
- https://www.mas.gov.sg/regulation/notices/notice-cmg-n03
- https://www.apra.gov.au/operational-risk-management

## Citation

- APA: Victoria Ashworth, LL.M, CAMS. (2026). Singapore MAS TRM & Notice PSN05/644 vs Australia APRA CPS 230: APAC Operational Resilience Map. RegTech Codex. https://regtechcodex.org/briefings/mas-trm-guidelines-apra-cps-230-apac-operational-resilience-map
- Canonical article: [Singapore MAS TRM & Notice PSN05/644 vs Australia APRA CPS 230: APAC Operational Resilience Map](https://regtechcodex.org/briefings/mas-trm-guidelines-apra-cps-230-apac-operational-resilience-map)
- Site research index: [RegTech Codex](https://regtechcodex.org/llms-full.txt)
