Google Security Operations is a cloud service that Google runs on its own infrastructure. nano is a SIEM you install on your servers, in your cloud account or air-gapped, licensed per GB/day.
| nano | Google SecOps | |
|---|---|---|
| What it is | A SIEM you install: log collection, OCSF parsing, nPL search, detections, cases, risk-based alerting and AI-assisted investigation in one product. docs · Getting started ↗ | "A cloud service, built as a specialized layer on top of Google infrastructure." It inherits that infrastructure's compute, storage and security design. docs.cloud.google.com · SecOps overview ↗ |
| Who operates it | You do, on your servers, in your own cloud account in any region, on Kubernetes with Helm, or air-gapped. nano, LLC never receives your logs. docs · Kubernetes ↗docs · Air-gapped ↗ | Google, as a cloud service. A lightweight forwarder can run in your network to send data to it. docs.cloud.google.com · SecOps overview ↗ |
| Collection | HEC, HTTP, OTLP, cloud pull (AWS S3 and SQS, GCP Pub/Sub, Azure Event Hubs, Kafka) and Vector agents. Every install starts with zero log sources; you choose what to onboard. docs · Integrations ↗ | "Data is ingested into the platform using forwarders, parsers, OpenTelemetry collectors, connectors, and webhooks." docs.cloud.google.com · SecOps overview ↗ |
| Data model | Parsers map each record to OCSF 1.9 at ingest, and nano stores the events in ClickHouse. docs · OCSF schema ↗ | Data is "normalized using the Unified Data Model (UDM), and linked to detections and threat intelligence." docs.cloud.google.com · SecOps overview ↗ |
| Search and detection language | nPL, nano's own query language, for search, dashboards and detections. Detections can live in Git as YAML files. docs · Porting rules ↗ | YARA-L 2.0, which Google describes as the query language powering Google SecOps for search, dashboards and rule-based threat detection. docs.cloud.google.com · YARA-L 2.0 ↗ |
| Investigation and response | Cases, playbooks and incidents, risk-based alerting, and pivt AI investigation on the model you choose. docs · Risk-based alerting ↗docs · pivt AI ↗ | Threats are investigated through case management, search and collaboration; analysts respond with automated playbooks and case management. docs.cloud.google.com · SecOps overview ↗ |
| Retention | As long as your storage allows. Older ClickHouse data can tier to your own S3-compatible object storage and stays searchable. docs · Storage ↗ | "By default, Google retains 12 months of your data." The period can be extended up to five years as part of your purchase order. docs.cloud.google.com · data retention ↗ |
The license is priced per GB/day of ingest. Nothing per user, endpoint or query.
Above 100 GB/day, sized with sales. Prices in US dollars, excluding taxes.
Every paid nano license includes all of this, at every GB/day size, on top of ingest, OCSF parsing, nPL search, detections and alerting.
| Capability | What it does |
|---|---|
| Cases, playbooks and incidents | Group related alerts into cases with evidence, timeline, queues and routing; correlate cases into incidents. |
| Risk-based alerting | Detections add risk to a user, host, IP or account; a risk rule alerts when the accumulated score crosses a threshold. docs · Risk-based alerting ↗ |
| pivt AI investigations | AI triage and investigation, off by default. When on, it calls the model endpoint you configure, including a self-hosted one. docs · pivt AI ↗ |
| SSO | OIDC single sign-on with group mapping and SCIM. |
| High availability and Kubernetes | Multi-node clusters and Helm deployment at every license size. docs · Kubernetes ↗ |
| Air-gapped operation | Offline license, signed image bundle and content packs you import by hand. docs · Air-gapped ↗ |
| Support | Business-hours support from nano. |
nano's engine is also free and open source under AGPL-3.0, with fewer features: open source.
What each familiar Google SecOps concept is called in nano, and what it does there.
| nano | Google SecOps | What it is in nano |
|---|---|---|
| OCSF, at ingest | UDM docs.cloud.google.com · UDM ↗ | nano's parsers map each record to OCSF 1.9 before it is stored. docs · OCSF schema ↗ |
| Log source and parser | Parser docs.cloud.google.com · UDM ↗ | Pick a parser from the catalog for each source, or have the AI parser builder write one from sample lines. docs · Integrations ↗ |
| nPL | UDM search, YARA-L 2.0 docs.cloud.google.com · YARA-L 2.0 ↗ | One pipe-based language for search and detection. docs · nPL ↗ |
| Detection rule | YARA-L rule docs.cloud.google.com · YARA-L 2.0 ↗ | An nPL rule over OCSF events that moves from Staging to Live to Alerting as you gain confidence in it. A coding agent ports your rules; you review. docs · Porting rules ↗ |
| Case | Case docs.cloud.google.com · cases ↗ | Groups related alerts with evidence, entities and a timeline; related cases correlate into incidents. docs · Cases ↗ |
Run both side by side, move sources over one at a time, and compare detections before you switch.
| Piece | What to do | Status |
|---|---|---|
| Sources | Onboard each source in nano from the catalog: every install starts with zero log sources. Google Cloud logs can arrive through a Pub/Sub pull subscription; Windows, Linux and syslog through Vector agents; applications through HEC or OTLP. | Per source docs · GCP Pub/Sub ↗docs · Integrations ↗ |
| Running both | Keep Google SecOps running while you compare. Where a collector can write to more than one destination, send the same events to both during the overlap. | Your plan |
| Rules and searches | A coding agent with nano's detection-as-code setup ports each rule to nPL from the source rule, and you review it before you enable it. | Agent-assisted docs · Porting rules ↗ |
| Historical data | Keep it in Google SecOps for its retention period, and plan the cut-over date around your retention obligations. | Your plan |
Size the target with the host sizing guide and check what your install sends to nano before you start.
Google Security Operations (Google SecOps) is Google's security operations platform. Google describes it as a cloud service built as a specialized layer on top of Google infrastructure. Data arrives through forwarders, parsers, OpenTelemetry collectors, connectors and webhooks, is normalized to Google's Unified Data Model (UDM) and linked to detections and threat intelligence, and is investigated through case management, search and playbooks.
Google describes Google SecOps as a cloud service on Google infrastructure. You can run a lightweight forwarder in your own network to send data to it, but the service itself is operated by Google. If the SIEM itself needs to run on your own servers, in another cloud account or air-gapped, nano is built for that.
Google's documentation says that by default Google retains 12 months of your data in your Google SecOps account, and that you can extend the retention period for up to five years as part of your purchase order.
YARA-L 2.0. Google describes it as the query language powering Google SecOps for search, dashboards and rule-based threat detection. nano uses its own language, nPL, for the same jobs.
Hand them to a coding agent with nano's detection-as-code setup. It ports each rule to nPL from the source rule and a nano example, and you review it before you enable it. Because nano normalizes to OCSF at ingest, a ported rule can cover other sources that map to the same fields; test it against each one before you enable it.
Yes. Cloud Logging's Log Router exports logs to a Pub/Sub topic, and nano reads them through a pull subscription, for example Cloud Audit Logs, VPC Flow Logs, Security Command Center findings and GKE logs. Each source is onboarded on purpose: every install starts with zero log sources.
Google SecOps is a cloud service operated by Google, built on UDM and YARA-L. nano is software you run yourself: events are normalized to OCSF at ingest and stored in ClickHouse, searched with nPL, and investigated with cases, risk-based alerting and pivt AI on your own model. It is licensed per GB/day with every feature included.
Nowhere outside your environment. nano is software you run on your own servers, in your own cloud account, on Kubernetes or air-gapped, and nano, LLC never receives your logs. A connected install sends nano a daily license check-in and one GB/day number, and downloads content and feeds; air-gapped installs send nothing. pivt AI is off by default and calls the model endpoint you configure.
Last reviewed 14 November 2026.
From US$149/month at 5 GB/day, US$279/month at 10 GB/day (list price, billed monthly; 10% off billed annually). Every feature in one license, and one license covers two installs.