Performance & latency

We take latency very seriously, and have architected our infrastructure to minimize it.

Our infrastructure

OpenPixels runs on a globally distributed set of edge workers. This architecture ensures our service is physically close to your servers, regardless of their location.

Expected performance

• We expect to add approximately 30ms of additional latency to each request.
• The first request to a new worker may take 100-200ms longer, as the worker loads your account information.

Connection pooling

The OpenPixels client maintains a persistent pool of connections to our servers. This approach eliminates the overhead of establishing new connections for each request, further reducing latency by avoiding TCP handshaking and TLS negotiation costs.

To take advantage of connection pooling, be sure to use the same client object for all of your requests.

Performance monitoring

If your requests are taking longer than expected, please let us know and we’ll investigate the issue.