☁️ Wednesday Noon: Architecture Vocabulary
“Great architecture is invisible — until it breaks.” Today we explore the language of cloud infrastructure and system design — words you’ll use in technical docs, stakeholder meetings, and architecture reviews.
🌟 Word of the Day
Scalability /ˌskeɪ.ləˈbɪl.ɪ.ti/
Vietnamese meaning: Khả năng mở rộng (của hệ thống)
The ability of a system to handle increasing amounts of work by adding resources.
3 Example Sentences:
- “We need to design the payment service with scalability in mind — we expect 10x traffic during the holiday season.”
- “One of the key advantages of microservices is scalability: you can scale individual components independently.”
- “The CTO asked us to present a scalability plan before we launch the new feature to all users.”
🔗 References:
📋 Vocabulary Table
| Phrase | IPA | Vietnamese Meaning | Example |
|---|---|---|---|
| load balancer | /ləʊd ˈbæl.ən.sər/ | Bộ cân bằng tải | ”We put a load balancer in front of our web servers to distribute traffic evenly.” |
| latency | /ˈleɪ.tən.si/ | Độ trễ | ”High latency is causing poor user experience in Southeast Asia — we need a CDN.” |
| fault tolerance | /fɔːlt ˈtɒl.ər.əns/ | Khả năng chịu lỗi | ”The system has built-in fault tolerance, so a single server failure won’t take down the app.” |
| throughput | /ˈθruː.pʊt/ | Thông lượng (xử lý) | “We increased throughput by 40% after optimizing our database query patterns.” |
| redundancy | /rɪˈdʌn.dən.si/ | Dự phòng (hệ thống) | “We added redundancy by deploying across three availability zones.” |
🗣️ Pronunciation Guide
Practice Sentence:
“Our load balancer improves latency and ensures fault tolerance across regions.”
Breakdown:
- load balancer → “LOHD BAL-en-ser” (stress on first syllable of each word)
- latency → “LAY-ten-see” (3 syllables, stress on LAY)
- fault tolerance → “FAWLT TOL-er-ents” (stress on FAWLT and TOL)
- regions → “REE-junz” (2 syllables, soft -juns ending)
Full sentence rhythm:
“Our LOHD BAL-en-ser im-PROOVZ LAY-ten-see and en-SHURZ FAWLT TOL-er-ents a-KROSS REE-junz.”
🔗 Audio References:
✏️ Exercise 1: Vocabulary in Context
Fill in the blanks with the correct word: (scalability / load balancer / latency / fault tolerance / redundancy)
- “Users in Europe are complaining about slow response times. We should investigate the _______ in our transatlantic connections.”
- “If one data center goes down, our _______ architecture ensures the app stays online.”
- “We’re expecting rapid growth next quarter, so _______ is a top architectural priority.”
- “The _______ distributes incoming API requests across five backend servers.”
- “We have _______ built in — every critical component has a backup running in parallel.”
✅ Check Your Answers
- latency — high latency = slow response time between distant regions
- fault tolerance — the system continues working despite failures
- scalability — priority when expecting growth/high traffic
- load balancer — distributes traffic across multiple servers
- redundancy — having backup components running in parallel
✏️ Exercise 2: Translation Challenge
Translate these Vietnamese sentences into professional English:
- “Hệ thống của chúng ta cần có khả năng mở rộng để xử lý lượng người dùng tăng gấp đôi vào năm sau.”
- “Độ trễ cao đang ảnh hưởng xấu đến trải nghiệm người dùng ở khu vực châu Á.”
- “Chúng ta nên thêm tính năng dự phòng để tránh single point of failure.”
✅ Suggested Translations
-
“Our system needs to be scalable to handle double the number of users by next year.” (Or: “We need to build scalability into our system to support 2x user growth next year.”)
-
“High latency is negatively impacting user experience in the Asia-Pacific region.” (Or: “Elevated latency is degrading UX for our users in Asia.”)
-
“We should add redundancy to avoid a single point of failure.” (Or: “We need to introduce redundancy into the architecture to eliminate any single point of failure.”)
💡 Idiom of the Day
”The single point of failure”
Vietnamese meaning: Điểm yếu duy nhất có thể làm sụp đổ cả hệ thống
Used in architecture discussions to describe a component that — if it fails — brings down the entire system. Also used metaphorically in business/team contexts.
Usage Examples:
-
“Right now, our database is a single point of failure. If it goes down, the whole app is offline. We need to add a read replica.”
-
“In this project, John is the single point of failure — he’s the only one who understands the legacy authentication system. We should document his knowledge.”
💬 Metaphorical use: You can use this phrase in business meetings to describe over-reliance on one person, process, or vendor — not just systems!
🎭 Mini Dialogue
Context: A backend engineer (Mai) explains system architecture to a non-technical product manager (Tom) before a stakeholder meeting.
Tom: “Mai, can you explain the new infrastructure in simple terms before we present to the board?”
Mai: “Sure! Think of it this way — we have a load balancer that acts like a traffic cop. It directs user requests to whichever server is least busy.”
Tom: “OK, and what about when servers crash?”
Mai: “That’s where fault tolerance comes in. We have redundancy built in — if one server fails, traffic automatically reroutes to a healthy one. Zero downtime.”
Tom: “What about performance? The board will ask about slowness.”
Mai: “We’ve cut latency by 60% using a CDN. Users in Vietnam now get responses in under 100 milliseconds.”
Tom: “And if we grow 10x next year?”
Mai: “That’s exactly why we designed for scalability. We can spin up new servers in minutes on AWS. The architecture scales horizontally.”
Tom: “Perfect. You just made me sound very smart in that meeting!”
🚀 Today’s Challenge
⏱ 2 minutes — right now:
Open Slack, your email, or a document you’re working on. Find ONE technical concept you’re discussing (a feature, a system, a service).
Write 2 sentences describing it using today’s vocabulary:
- One sentence about its scalability or fault tolerance
- One sentence about its latency or throughput
💬 Example: “The new notification service is designed for scalability — it can handle 1M events per minute. However, we’re still optimizing latency; currently p99 response time is 250ms, and we’re targeting under 100ms.”
Bonus: Say the two sentences out loud three times — focus on stress patterns!
📊 Today’s Session Summary
| Skill | What You Practiced |
|---|---|
| 🗣️ Pronunciation | latency, scalability, fault tolerance, load balancer |
| 📖 Vocabulary | 5 core architecture terms with IPA + examples |
| ✍️ Writing | Translation & gap-fill exercises |
| 💬 Speaking | Mini dialogue + challenge sentence |
| 🧠 Idiom | ”single point of failure” — technical & metaphorical |
💡 Come back at 6:00 PM for the Evening Session — we’ll focus on applying today’s vocabulary in written communication: emails, Slack messages, and technical docs.
🔁 Also: Check this morning’s Morning Session on Architecture phrases for non-tech stakeholders!