Comment on page
Our service API Latency
Latency is very important to us and we constantly monitor performance of our API to ensure our world class products have high performance.
We consistency implement changes to maintain and improve latency performance of our services.
When calling our API service, several factors can affect end user latency, which is the delay between making a request and receiving a response. Its important to understand these factors
- 1.Network Latency:
- Physical distance: The geographical distance between the client and the API server can significantly affect latency.
- Network congestion: High traffic and congestion on the network can introduce delays in data transmission.
- Network latency is outside of our control.
- 2.GBG On-Server Performance:
- Server load: The server's current load or the number of requests it's processing can affect response times.
- We manage this by great api design, auto scaling our servers during peak demand periods..
- 3.Client-Side Factors:
- Client hardware and software: The performance of the client device, as well as the efficiency of the client application, can influence latency.
- Client Server load: The server's current load or the number of requests it's processing can affect response times.
- Client security protocols.
- Client-side factors are outside of our control.
- 4.Authentication and Security:
- Encryption: We use TLS of encryption, can add processing time to both the client and server, affecting latency.
- This is essential for information security.
- 5.Auto Scaling delay:
- We auto scale services to demand levels to manage traffic and latency.
- In peak demand, it takes approximately 30 -120 seconds for the servers to scale up to meet maintain our target On-Server latency. During this short auto scaling period we see peak latency.
- We have short term request caching on Server-side to reduce overall latency.
- 7.Third-Party Dependencies:
- Some API services relies on third-party services or external data sources, the latency of those services can impact the overall API response time.
- 8.Datasets accessed:
- Latency varies between the datasets by accessed.
- With AU & NZ datasets this is tiny 20ms. Larger impact is when accessing Non AU & NZ addresses.
We measure the Address Lookup service (Address Find) because this is user typeahead services. 2023 AU & NZ - Average On-Server Latency is 42ms.
2023 INTL - Average On-Server Latency is 449ms. Last month average was 378ms. In 2023, we have reduced latency 39%. We continue to work on reducing latency of this service.
2023 AU & NZ - Average On-Server Latency is 20ms.
2023 INTL - Average On-Server Latency is 86ms.
2023 - Average On-Server Latency is 3ms.
2023 AU & NZ - Average On-Server Latency is 32ms.
2023 INTL - Average On-Server Latency is 466ms. Last month average was 314ms. We continue to work on reducing latency of this service targeting sub 100ms.
2023 AU & NZ - Average On-Server Latency is 14ms.
2023 Average On-Server Latency is 58ms.
2023 Average On-Server Latency for is 534ms (Oct).
We have implement a timeout feature on email validation. This service is heavy reliant (industry issue) on 3rd party services such network latency, email server latency and responses.
We implement change Sept 2023 which has reduced latency by 50%.
2023 Average On-Server Latency for is 1992ms.
This service is heavy reliant (industry issue) on 3rd party services such network latency and live phone connectivity tests.
2023 Average On-Server Latency for is 234ms.