How to Talk About System Capacity and Scalability in English at Work
ESL English learning: Master capacity and scalability vocabulary for system design discussions, architecture reviews, and explaining constraints to stakeholders.
Pronunciation Guide: Capacity Terms That Trip People Up
These terms appear constantly in architecture discussions. Mispronouncing them can undermine your credibility, even if your technical knowledge is solid.
scalability /ˌskeɪləˈbɪləti/
ska-la-BI-li-ty (stress on third syllable)
Saying 'scale-A-bility' with stress on second syllable
throughput /ˈθruːpʊt/
THROUGH-put (stress on first syllable)
Pronouncing 'th' as /t/ or /s/; saying 'through-PUT'
bottleneck /ˈbɒtlnek/
BOT-tle-neck (stress on first syllable)
Saying 'bottle-NECK' with stress on last syllable
latency /ˈleɪtənsi/
LAY-ten-cy (stress on first syllable)
Saying 'la-TEN-cy' with stress on second syllable
utilization /ˌjuːtɪlaɪˈzeɪʃən/
u-ti-li-ZAY-shun (stress on fourth syllable)
Saying 'u-TI-li-zation' with stress on second syllable
Why Capacity Vocabulary Matters in Tech
Whether you're writing a design doc, explaining constraints to a PM, or reviewing infrastructure during an incident, you need precise vocabulary to describe system capacity. Vague phrases like 'the database is getting full' don't give stakeholders actionable information.
This article maps essential capacity and scalability vocabulary to real tech workplace scenarios — from Slack threads to architecture reviews. You'll learn not just what these terms mean, but how to use them naturally in spoken and written English.
Culture Note: Precision vs. Panic
In tech culture, saying 'we're running out of space' sounds alarming. Saying 'we're at 82% storage utilization with 3 weeks of headroom at current growth rates' shows you understand the situation and have measured it. Precise vocabulary builds trust.
Vocabulary for Systems with High Capacity
When systems have plenty of capacity, we use vocabulary that conveys room to grow. These terms help you communicate positive capacity states in design docs and status updates.
| Term | Tech Definition | Example in Context |
|---|---|---|
| headroom | Available capacity above current usage for handling spikes | We have 40% headroom on the primary database cluster. |
| ample | More than enough capacity for current and near-future needs | The new storage tier provides ample room for the next two quarters. |
| scalable | Designed to handle increased load by adding resources | The microservices architecture is highly scalable. |
| elastic | Able to automatically expand and contract based on demand | Our elastic compute layer scales up during peak hours. |
| extensive | Covering a wide range of capacity or functionality | AWS offers extensive options for storage tiers. |
- Use 'headroom' when quantifying buffer capacity: 'We need at least 30% headroom for Black Friday traffic.'
- Use 'ample' in positive assessments: 'The migration gives us ample capacity for the product roadmap.'
- Use 'elastic' specifically for auto-scaling systems, not manual scaling.
Vocabulary for Capacity Constraints
When systems are strained, you need vocabulary that precisely describes the constraint without causing unnecessary alarm. These terms help you write clear incident reports and explain limitations to non-technical stakeholders.
| Term | Tech Definition | Connotation |
|---|---|---|
| bottleneck | A component limiting overall system throughput | Neutral/technical — identifies the constraint point |
| saturated | Operating at or near maximum capacity | Technical — implies metrics are at ceiling |
| cramped | Insufficient resources for comfortable operation | Slightly informal — suggests discomfort |
| congested | Too many requests competing for limited resources | Neutral — often used for network/queue issues |
| constrained | Limited by a specific factor | Formal — good for docs and executive summaries |
| tight | Very little margin or headroom remaining | Informal — common in Slack and standups |
Grammar Note: Bottleneck as Noun and Verb
Bottleneck works as both noun ('The database is the bottleneck') and verb ('Database queries are bottlenecking the checkout flow'). The verb form is common in spoken English but less formal in writing.
Horizontal vs. Vertical Scaling: Getting the Terms Right
Scaling vocabulary is essential for architecture discussions. Using these terms correctly shows you understand infrastructure trade-offs and can communicate them clearly.
| Vertical Scaling (Scale Up) | Horizontal Scaling (Scale Out) |
|---|---|
| Add more CPU, RAM, or storage to existing machines | Add more machines or nodes to distribute load |
| 'We scaled up the database server to handle the query load.' | 'We scaled out to five read replicas across regions.' |
| Simpler but has hard limits | More complex but theoretically unlimited |
| Often a short-term fix | Usually a long-term architectural choice |
- Say 'scale up/down' for vertical scaling: 'We can scale up the instance type if queries slow down.'
- Say 'scale out/in' for horizontal scaling: 'The system scales out automatically during traffic spikes.'
- Avoid mixing metaphors: don't say 'scale up more servers' (that's scaling out).
Slack vs. Design Doc: Register Differences
How you describe capacity issues changes based on context. A Slack message to your team can be informal, but a design doc or RFC needs precise, formal language. Here's how the same information sounds in different registers.
| Informal (Slack/Standup) | Formal (Design Doc/RFC) |
|---|---|
| DB is getting pretty cramped — we're at like 85% | The primary database is at 85% storage utilization with limited headroom for growth. |
| We're bottlenecked on the auth service right now | The authentication service is currently the primary throughput constraint. |
| We can just throw more boxes at it | We can address this through horizontal scaling of the compute tier. |
| Storage is tight, maybe 2 weeks before it's a problem | At current ingestion rates, we will exhaust available storage capacity within approximately 14 days. |
| The queue is getting backed up pretty bad | The message queue is experiencing significant congestion due to consumer lag. |
Culture Note: Know Your Audience
With your engineering team, informal language builds rapport. With executives or in written documentation, formal precision shows professionalism. Most tech workers code-switch between these registers multiple times per day.
Common Mistakes: What Engineers Actually Say vs. Better Alternatives
These are real patterns heard in tech workplaces. The 'heard at work' versions are understandable but imprecise or awkward. The 'better English' versions are clearer and more professional.
| Heard at Work ❌ | Better English ✓ |
|---|---|
| The system has no more space for data. | The system is approaching storage capacity limits. |
| We need to make the database bigger. | We need to scale up the database instance / expand storage allocation. |
| There is a problem with too many requests. | The service is experiencing request congestion / is saturated. |
| We should add more capacity before the system breaks. | We should provision additional headroom before utilization reaches critical thresholds. |
| The network is full. | The network is congested / bandwidth is saturated. |
| We scaled up three new servers. | We scaled out by adding three new servers. (Scale up = bigger machine, scale out = more machines) |
Practice Exercises
Test your understanding of capacity and scalability vocabulary with these exercises based on real workplace scenarios.
Multiple choice
Choose the best answer.
Your PM asks why the checkout page is slow. Which response uses the correct technical vocabulary?
Multiple choice
Choose the best answer.
Complete this design doc sentence: 'We recommend maintaining at least 30% _______ on the primary cluster to handle unexpected traffic spikes.'
Multiple choice
Choose the best answer.
Which sentence correctly describes horizontal scaling?
Multiple choice
Choose the best answer.
In a formal RFC, which phrasing is most appropriate for describing a capacity problem?