Skip to content
This repository was archived by the owner on May 9, 2026. It is now read-only.

Performance

Nathanial Hwong edited this page Apr 1, 2026 · 8 revisions

1. Improve Claims List Rendering with FlashList

Replace the existing ScrollView/FlatList in the claims dashboard with @shopify/flash-list to enhance performance when rendering multiple claim cards. FlashList virtualizes list items, rendering only the visible items plus a small buffer, which reduces memory usage and improves scrolling smoothness.

Code Implementation

View changes in commit

2. System Performance Estimations

Scope

This page estimates daily workload and baseline response-time characteristics for AERO based on current code and deployment settings.

Notes on our system

  • Flight data is currently being fetched from 3 different airports (YUL, YVR, YYZ).
  • Data is fetched once every 3 hours.

Daily flight processing estimate

1194 flights daily × 35% of flights delayed × 8 daily fetches = 3623 flights processed/day

Note: An important assumption was made regarding delayed flight data retrieval. We assumed that the API endpoints return delays and cancellations covering approximately the past 24 hours; however, the actual time window and potential lag are not explicitly defined and may vary.


Daily claims processed estimate

  • Using previously researched data, we have 452 flights delayed per day from YUL, YYZ, and YVR collectively
  • Rough assumption: 150 passengers per flight. (A mix of different aircrafts depart from these airports, most common capacities are: 80, 160, 300 with smaller flights being more common)

Daily claims will depend heavily on adoption by the public. We can calculate the estimated number of claims based on a spread of different possible adoption rates. (A combination of factors have influenced to believe only a small portion of passengers will be likely to submit claims. First, there are a small number of already present competitors. Second, passengers are generally unlikely to submit claims due to lack of awareness of compensation regulation which can be improved hence the wide range of adoption rates considered).

  • Low adoption: 0.1%
  • Medium adoption: 0.8%
  • High adoption 3%

Low: 0.001 * 452 * 150 = 68

Medium: 0.01 * 452 * 150 = 543

High: 0.03 * 452 * 150 = 2034

Even when assuming only 1 out of 1000 passengers submit a claim to our system, we still average 68 requests per day.


Clone this wiki locally