Event schema

Fields the query expects

Field	Type	Why it exists
timestamp_utc	Timestamp	Resource completion time.
resource_host	Utf8	Approved hostname without path or query values.
resource_type	Utf8	script, stylesheet, image, fetch, or another bounded class.
duration_ms	Float64	Resource duration.
transfer_bytes	Int64	Transferred response bytes when available.
status	Utf8	success, failed, or blocked.

DataFusion SQL

Copy the query

sql

SELECT
  resource_host,
  resource_type,
  COUNT(*) AS resources,
  approx_percentile_cont(duration_ms, 0.95) AS p95_duration_ms,
  SUM(transfer_bytes) AS transfer_bytes,
  SUM(CASE WHEN status <> 'success' THEN 1 ELSE 0 END) AS failures
FROM frontend_resource_events
WHERE timestamp_utc >= now() - INTERVAL '24 hours'
GROUP BY resource_host, resource_type
HAVING COUNT(*) >= 20
ORDER BY p95_duration_ms DESC, failures DESC;

This read-only query is planned and executed against an empty typed table with Apache DataFusion 45.2.0. We review the synthetic sample output separately. Check field types, thresholds, and counting rules against your own data. Read the testing methodology.

Query result

p95 resource duration by host

The analytics script is slower and fails more often than the first-party fetch group.

Compare a numeric result field

analytics.example

1,820 ms

comparison 42 ms

api.example

910 ms

comparison 18 ms

resource_host	resource_type	resources	p95_duration_ms	transfer_bytes	failures
analytics.example	script	2,140	1,820	68,480,000	42
api.example	fetch	8,210	910	124,200,000	18

Synthetic example output. Run the query against your own event schema and thresholds before using it for operational decisions.

p95 resource duration by host: static chart of synthetic p95_duration_ms values from the Find slow frontend resources by host example result — Download this SVG chart of the sample results for an article, runbook, or design review. Please credit Telemetry.

Reproduce the example

Download the sample data

The JSON bundle includes the event schema with field types, illustrative input rows, exact SQL, expected output, review notes, and engine version. The CSV contains the displayed result.

Sample JSON Expected output CSV Static SVG chart

How the SQL works

1Host and type preserve ownership without storing full resource URLs.
2High p95 duration shows that some third-party requests are slow even when most finish sooner.
3Compare transferred bytes with duration to check whether larger responses explain the slowdown.

Edge cases to check

Cross-origin timing detail depends on browser policy.
Cached resources require separate interpretation.
Never store signed URLs, query strings, or user-specific paths.

Recommended dashboard

Bars: p95_duration_ms by host
Trend: failures by resource type
Table: transfer_bytes by approved host

Alert guidance

Alert when a critical first-party host fails or a reviewed third-party p95 crosses its budget.

Read alert setup

Set up the events this query needs

Related instrumentation and guides

Frontend performance monitoring Frontend reliability recipes API latency percentiles

Continue the analysis

Run it on your events

Create a table, adapt the fields, and save the result

Start free, send structured events, and use the query result as a chart, shared dashboard widget, or alert input.

Get an API key

Find slow frontend resources by host