التوافر العالي والتعافي من الكوارث
في هذه الصفحة 9 موضوعات
نص الدروس والتمارين باللغة الإنجليزية، بينما تبقى واجهة الموقع وعناوين الدروس بالعربية.
نظرة عامة
الموضوعات
- تعدد مناطق التوافر (Multi-AZ)
- تعدد المناطق (Multi-region)
- RPO/RTO
- تصميم التعافي من الكوارث
تعلّم أولًا
4 ساعة تدريب (تقدير)
مخرجات التعلم
- Translate business impact into RPO and RTO per capability.
- Choose multi-AZ and multi-region patterns (backup-restore, pilot light, warm standby, active-active) to meet them at acceptable cost.
- Design DR for dependencies the solution doesn't own (ERP, payment provider, identity).
- Define DR testing and evidence.
الدروس
- 01التوافر والتعافي كقرارات عمل
- 02تصميم Multi-AZ
- 03أنماط التعافي من الكوارث، من الاستعادة من النسخ الاحتياطي إلى النشط-النشط
- 04اعتماديات لا تملكها
- 05اختبار التعافي من الكوارث
- 06عرض عملي: تقدير تكلفة ثلاثة أنماط للتعافي من الكوارث لعملية الدفع في Larkspur
- 07عرض عملي: ماذا يحدث للمتاجر عندما تتعطل المنطقة السحابية
المختبر والتمارين
Lab
ARC-019-LAB1 · Larkspur HA/DR design
Outcomes: 1–4 · Time: ~4 h
- List Larkspur capabilities (checkout, catalog, store tills, fulfillment, reporting, back office) and their business impact per hour.
- Set RPO and RTO per capability with the sponsor's assumptions.
- Choose the multi-AZ design and the DR pattern per capability and cost each option.
- Document fallbacks for the ERP, payment provider, and identity, including stores selling offline.
- Write a DR test plan with scenarios, success criteria, and evidence.
Acceptance: targets met at a cost the sponsor accepts, or trade-off escalated.
Exercises (case-based; six exercises)
| ID | Outcome | Task |
|---|---|---|
| ARC-019-EX01 | 1 | Convert these impacts into RTO and RPO tiers: checkout loses 8,000 per hour, reporting loses little, the loyalty app loses goodwill. |
| ARC-019-EX02 | 1 | The sponsor says "everything must be five nines". Prepare the conversation. |
| ARC-019-EX03 | 2 | Pick a DR pattern for checkout, catalog, and reporting and justify each. |
| ARC-019-EX04 | 2 | Describe what breaks if the design uses three zones but only enough capacity for three zones' load. |
| ARC-019-EX05 | 3 | The payment provider has an outage. Define the degraded behavior and customer communication. |
| ARC-019-EX06 | 4 | Write a DR test plan for the checkout database failover with success criteria. |
أظهر الحلول
- EX01 Checkout: RTO under one hour, RPO minutes; catalog: RTO a few hours, RPO hours; reporting: RTO a day, RPO a day; loyalty: relaxed but monitored. Confirm with the sponsor.
- EX02 Show cost per extra nine, ask what an hour of outage costs, propose tiered targets, make the sponsor choose with prices.
- EX03 Checkout: warm standby; catalog: pilot light or restore with CDN cache; reporting: backup-restore.
- EX04 Losing one zone leaves two thirds of capacity for full load, so users see slowness or errors; size with N+1 zones.
- EX05 Offer alternate payment methods if contracted, queue orders for later capture where legal and safe, show clear messages, and track the provider's recovery.
- EX06 Scenario: primary zone failure and region failure; steps; measure time to recover and data loss; success: within target RTO and RPO; record results, findings, owners, next test date.
Workplace evidence: an HA/DR design you produced and tested.
Outcome coverage: all outcomes covered by exercises and the lab.
الاختبار
Outcomes: O1 RPO and RTO · O2 multi-AZ and DR patterns · O3 dependencies · O4 DR testing.
اختبار ذاتي لك: لا توجد درجات، ولا يُحفظ أو يُرسل أي شيء. اختر إجابة ثم تحقق منها، ويمكنك المحاولة مرة أخرى كما تشاء.
الجزء العملي
HA/DR design defense using the ARC-019-LAB1 design, the costed options, and exercises ARC-019-EX01 to EX06. Pass: quiz at least 8/10, at least five of six exercises correct, and targets met at an accepted cost or the trade-off escalated.
أخطاء شائعة
One RPO/RTO for everything; DR plans that ignore the ERP and payment provider.
التقييم
HA/DR design defense.
دليل من بيئة العمل
An HA/DR design you produced and tested.
الصلة بالمسار الوظيفي
Resilience architecture (ARC-031).