Log analytics with no index tax.
NIXL stores compact columnar data in S3 and scans it fast, instead of building heavy indexes on the way in. You pay for the data a scan reads — not the terabytes you store.
Built on AWS · KQL today, SQL coming · billed on data scanned
Skip the index. Scan instead.
Index-first tools pay a tax on every byte you ingest: build the index, store it, keep it warm. NIXL takes the other path — compact columnar data with lightweight skip-indexes, scanned by a vectorized engine only when you query.
No inverted index on ingest
Columnar data plus skip-indexes (zone maps, bloom filters) in S3 — not a full inverted index built on the way in.
Pay for what you query
Cost tracks the data a scan reads, not what you ingest or retain. Keeping more data queryable stays cheap.
Two teams, one bill
DevOps and security query the same logs at scale — without standing up two tools and paying two bills.
Keep more data queryable, for less.
Debug across long time ranges without pre-deciding what to index. Because cost tracks usage rather than retention, you can keep weeks of logs hot and still query them when an incident actually needs them.
- / Query raw logs at full fidelity, not a sampled subset.
- / No index sizing or shard tuning to plan around.
- / KQL today, SQL as a fast-follow.
Investigate at full volume.
Hunt and investigate over the data you actually kept, not the slice you could afford to index. NIXL is log analytics, not a SIEM — it gives security teams fast search over large volumes of retained data, alongside the same engine the dev team uses.
- / Search across long retention windows without re-hydrating.
- / The full dataset is queryable, so investigations aren't blind to what was dropped.
- / Shared engine with DevOps — one source of truth for logs.
We'll lead with measured numbers, not adjectives.
A public, reproducible benchmark — including a head-to-head against Elastic — is the proof artifact. Until it's published, we won't print speed or cost multipliers. Sign up and you'll get it the day it lands.
Scan speed at scale vs index-first tools — published, reproducible.
Cost at real volumes, from primary-source pricing — published, dated.
Head-to-head against Elastic — fully published test.
Built by database and logging veterans who want to make DevOps and SecOps faster, cheaper and better.
Get notified when NIXL opens up.
Early access and the public benchmark, straight to your inbox. No spam — only NIXL availability.