Query response time degradation

I opened a ticket (#481667) about this issue last Friday, but haven’t seen a response yet, so am posting here as well. Apologies if I’m using the wrong support channels, and sorry for the duplicate message.

Over the last couple of days (graph from New Relic attached), we’ve seen a marked increase in response time for search queries made to Algolia. Some outliers are taking over a minute, leading to timeouts on our side and poor user experience (we proxy search requests through our server).

I don’t think we’ve changed anything appreciable about our requests or search index configuration recently. Is there any evidence of a performance degradation occurring on Algolia’s side, perhaps? I’ve looked at the Algolia performance metrics, and haven’t found a graph similar to our New Relic chart, but I’m wondering if there’s a routing layer or something not represented in the Algolia performance charts that could be the culprit.

Can you help us figure out root cause here? Thoughts and debugging tips appreciated!