For platform & observability teams

    Stop sampling your logs
    to save money.

    S3-compatible storage at $4.99/TB flat. No per-request fees, no egress. Keep every event, every span, every audit trail — without watching the PUT counter.

    Start for freeTalk to an expert

    No credit card required · No per-request fees · Connects in minutes

    Logs are billed per write. Logs are written constantly.

    Hyperscaler object storage charges per PUT. A logging pipeline writes by definition. The cheapest way to make the bill smaller is to keep fewer logs — and the price paid for that decision is paid later, in the incident postmortem you cannot reconstruct.

    What you write

    Each write is a metered event.

    100 million PUTs per month is a modest production logging workload. On AWS S3 that is $500 in request charges alone, before a single byte of storage is billed.

    What you store

    Storage is the smaller line.

    10 TB of compressed log data per month is normal for a mid-size platform. Add storage at $0.023/GB and egress for any query that reads back — the bill closes in on $750 a month before you draw a single dashboard.

    What you give up

    The bill picks which events to keep.

    Engineering reacts by sampling. Trace volume gets capped. Debug logs get truncated. The dashboard you finally build only has the events that survived the budget meeting.

    A logging sink that doesn't bill per event.

    Vector, Fluent Bit, Logstash, OpenTelemetry — anything that already writes S3 — gets a new endpoint. The PUT counter stops mattering.

    vector.toml
    # vector.toml — write every event to Fil One
[sinks.fil_one_logs]
type      = "aws_s3"
inputs    = ["all_services"]
bucket    = "prod-logs"
endpoint  = "https://eu-west-1.s3.fil.one"
region    = "eu-west-1"
compression = "gzip"

[sinks.fil_one_logs.batch]
max_bytes = 10485760    # flush 10 MiB at a time
timeout_secs = 60       # one PUT per shard per minute

[sinks.fil_one_logs.auth]
access_key_id     = "${FIL_ACCESS_KEY}"
secret_access_key = "${FIL_SECRET_KEY}"
    AWS S3 Standard$0.005 / 1K PUT$500.00
    Google Cloud$0.05 / 10K Class A$500.00
    Azure Blob$0.055 / 10K writes$550.00
    Wasabi$0 per request$0
    Backblaze B2$0 per request$0
    Fil One$0 per request$0

    Public US rate cards, Q2 2026. Storage and egress not included in this line — they are extra on the metered tiers, and zero on Fil One.

    A sink that scales with retention, not write rate.

    Same S3 API, same SDKs. The only thing that changes is the line item that used to dominate the bill.

    No per-request fees

    PUT, GET, LIST, HEAD — all included. The line item that dominates a logging workload on AWS does not exist here.

    S3-compatible logging

    Vector, Fluent Bit, Logstash, OpenTelemetry collectors, Loki — all of them write S3. Point them at the Fil One endpoint and ship.

    Predictable bill

    Storage charges grow with what you keep, not with how loudly your services log. A noisy deploy stops being a billing event.

    Versioning and Object Lock

    Tamper-evident audit logs by default. Compliance-mode retention for SOC 2, ISO 27001, and HIPAA evidence retention periods.

    One rate. $4.99/TB/month.

    Storage. That is the whole bill. No PUT charges. No egress. Logs cost what they should — the bytes you decide to keep.

    Start for freeTalk to an expert

    No credit card required · No per-request fees · Connects in minutes

    Keep every log.

    Free 1 TB evaluation. Point your existing collector at the endpoint and watch the request line zero out.

    Start for freeTalk to an expert

    No credit card required · No per-request fees · Connects in minutes