Add GCP self-hosted architecture documentation

Author: DavoCoder

src/docs.json

@@ -1341,7 +1341,8 @@
                   "group": "Self-hosted cloud architecture",
                   "pages": [
                     "langsmith/aws-self-hosted",
-                    "langsmith/azure-self-hosted"
+                    "langsmith/azure-self-hosted",
+                    "langsmith/gcp-self-hosted"
                   ]
                 },
                 {

src/langsmith/gcp-self-hosted.mdx

@@ -0,0 +1,109 @@
+---
+title: Self-hosted on GCP
+sidebarTitle: GCP
+icon: "google"
+---
+
+When running LangSmith on [Google Cloud Platform (GCP)](https://cloud.google.com/), you can set up in either [full self-hosted](/langsmith/self-hosted) or [hybrid](/langsmith/hybrid) mode. Full self-hosted mode deploys a complete LangSmith platform with observability functionality as well as the option to create agent deployments. Hybrid mode entails just the infrastructure to run agents in a data plane within your cloud, while our SaaS provides the control plane and observability functionality.
+
+This page provides GCP-specific architecture patterns, service recommendations, and best practices for deploying and operating LangSmith on GCP.
+
+<Note>
+LangChain provides Terraform modules specifically for GCP to help provision infrastructure for LangSmith. These modules can quickly set up GKE clusters, Cloud SQL, Memorystore Redis, Cloud Storage, and networking resources.
+
+View the [GCP Terraform modules](https://github.com/langchain-ai/terraform/tree/main/modules/gcp) for documentation and examples.
+</Note>
+
+## Reference architecture
+
+We recommend leveraging GCP's managed services to provide a scalable, secure, and resilient platform. The following architecture applies to both self-hosted and hybrid and aligns with the [Google Cloud Well-Architected Framework](https://docs.cloud.google.com/architecture/framework):
+
+![Architecture diagram showing GCP relations to LangSmith services](/langsmith/images/gcp-architecture-self-hosted.png)
+
+- <Icon icon="globe" /> **Ingress & networking**: Requests enter via [Cloud Load Balancing](https://cloud.google.com/load-balancing) within your [VPC](https://cloud.google.com/vpc), secured using [Cloud Armor](https://cloud.google.com/armor) and [IAM](https://cloud.google.com/iam)-based authentication.
+- <Icon icon="cube" /> **Frontend & backend services:** Containers run on [Google Kubernetes Engine (GKE)](https://cloud.google.com/kubernetes-engine), orchestrated behind the load balancer. Routes requests to other services within the cluster as necessary.
+- <Icon icon="database" /> **Storage & databases:**
+  - [Cloud SQL for PostgreSQL](https://cloud.google.com/sql/docs/postgres): metadata, projects, users, and short-term and long-term memory for deployed agents. LangSmith supports PostgreSQL version 14 or higher.
+  - [Memorystore for Redis](https://cloud.google.com/memorystore/docs/redis): caching and job queues. Memorystore can be in single-instance or cluster mode, running Redis OSS version 5 or higher.
+  - ClickHouse + [Persistent Disks](https://cloud.google.com/compute/docs/disks): analytics and trace storage.
+    - We recommend using an [externally managed ClickHouse solution](/langsmith/self-host-external-clickhouse) unless security or compliance reasons
+    prevent you from doing so.
+    - ClickHouse is not required for hybrid deployments.
+  - [Cloud Storage](https://cloud.google.com/storage): object storage for trace artifacts and telemetry.
+
+- <Icon icon="sparkles" /> **LLM integration:** Optionally proxy requests to [Vertex AI](https://cloud.google.com/vertex-ai) for LLM inference.
+- <Icon icon="chart-line" /> **Monitoring & observability:** Integrate with [Cloud Monitoring](https://cloud.google.com/monitoring) and [Cloud Logging](https://cloud.google.com/logging)
+
+
+## Compute options
+
+LangSmith supports multiple compute options depending on your requirements:
+
+| Compute option | Description | Suitable for |
+|-----------------|-------------|--------------|
+| **Google Kubernetes Engine (preferred)** | Advanced scaling and multi-tenant support | Large enterprises |
+| **Compute Engine-based** | Full control, BYO-infra | Regulated or air-gapped environments |
+
+## Google Cloud Well-Architected best practices
+
+This reference is designed to align with the six pillars of the Google Cloud Well-Architected Framework:
+
+### Operational excellence
+
+- Automate deployments with IaC ([Terraform](https://www.terraform.io/) / [Deployment Manager](https://cloud.google.com/deployment-manager)).
+- Use [Secret Manager](https://cloud.google.com/secret-manager) for configuration and sensitive data.
+- Configure your LangSmith instance to [export telemetry data](/langsmith/export-backend) and continuously monitor via [Cloud Logging](https://cloud.google.com/logging).
+- The preferred method to manage [LangSmith deployments](/langsmith/deployments) is to create a CI process that builds [Agent Server](/langsmith/agent-server) images and pushes them to [Artifact Registry](https://cloud.google.com/artifact-registry). Create a test deployment for pull requests before deploying a new revision to staging or production upon PR merge.
+
+### Security
+
+- Use [IAM](https://cloud.google.com/iam) roles with least-privilege policies and [Workload Identity](https://cloud.google.com/kubernetes-engine/docs/how-to/workload-identity) for secure pod-to-GCP-service authentication.
+- Enable encryption at rest ([Cloud SQL](https://docs.cloud.google.com/sql/docs/postgres/cmek), [Cloud Storage](https://cloud.google.com/storage/docs/encryption), Persistent Disks) and in transit (TLS 1.2+).
+- Integrate with [Secret Manager](https://cloud.google.com/secret-manager) for credentials.
+- Use [Identity Platform](https://cloud.google.com/identity-platform) or [Workload Identity Federation](https://cloud.google.com/iam/docs/workload-identity-federation) as an IDP in conjunction with LangSmith's built-in authentication and authorization features to secure access to agents and their tools.
+
+### Reliability
+
+- Replicate the LangSmith [data plane](/langsmith/data-plane) across regions: Deploy identical data planes to Kubernetes clusters in different regions for LangSmith Deployment. Deploy [Cloud SQL](https://cloud.google.com/sql/docs/postgres/high-availability) and [GKE](https://docs.cloud.google.com/kubernetes-engine/docs/concepts/configuration-overview) services across multiple zones.
+- Implement [autoscaling](https://cloud.google.com/kubernetes-engine/docs/concepts/cluster-autoscaler) for backend workers using [Horizontal Pod Autoscaler](https://cloud.google.com/kubernetes-engine/docs/concepts/horizontalpodautoscaler) and [Cluster Autoscaler](https://cloud.google.com/kubernetes-engine/docs/concepts/cluster-autoscaler).
+- Use [Cloud DNS](https://cloud.google.com/dns) health checks and failover policies.
+
+### Performance optimization
+
+- Leverage [Compute Engine](https://cloud.google.com/compute) instances for optimized compute with [machine type selection](https://cloud.google.com/compute/docs/machine-types).
+- Use [Cloud Storage lifecycle policies](https://cloud.google.com/storage/docs/lifecycle) for infrequently accessed trace data, moving to [Nearline](https://cloud.google.com/storage/docs/storage-classes#nearline) or [Coldline](https://cloud.google.com/storage/docs/storage-classes#coldline) storage classes.
+
+### Cost optimization
+
+- Right-size [GKE](https://cloud.google.com/kubernetes-engine) clusters using [Committed Use Discounts](https://cloud.google.com/compute/docs/instances/signing-up-committed-use-discounts) and [Sustained Use Discounts](https://cloud.google.com/compute/docs/sustained-use-discounts).
+- Monitor cost KPIs using [Cloud Billing](https://cloud.google.com/billing/docs) dashboards and [Cost Management](https://cloud.google.com/cost-management) tools.
+
+### Sustainability
+
+- Minimize idle workloads with on-demand compute and [autoscaling](https://cloud.google.com/kubernetes-engine/docs/concepts/cluster-autoscaler).
+- Store telemetry in low-latency, low-cost tiers using [Cloud Storage lifecycle policies](https://cloud.google.com/storage/docs/lifecycle).
+- Enable auto-shutdown for non-prod environments using [scheduled actions](https://cloud.google.com/compute/docs/instances/schedule-instance-start-stop).
+
+## Security and compliance
+
+LangSmith can be configured for:
+
+- [Private Service Connect](https://cloud.google.com/vpc/docs/private-service-connect)-only access (no public internet exposure, besides egress necessary for billing).
+- [Cloud KMS](https://cloud.google.com/kms)-based encryption keys for Cloud Storage, Cloud SQL, and Persistent Disks.
+- Audit logging to [Cloud Logging](https://cloud.google.com/logging) and [Cloud Audit Logs](https://cloud.google.com/logging/docs/audit).
+
+Customers can deploy in [Assured Workloads](https://cloud.google.com/assured-workloads) regions for compliance with ISO, HIPAA, or other regulatory requirements as needed.
+
+## Monitoring and evals
+
+Use LangSmith to:
+
+- Capture traces from LLM apps running on [Vertex AI](https://cloud.google.com/vertex-ai).
+- Evaluate model outputs via [LangSmith datasets](/langsmith/manage-datasets).
+- Track latency, token usage, and success rates.
+
+Integrate with:
+
+- [Cloud Monitoring](https://cloud.google.com/monitoring) dashboards.
+- [OpenTelemetry](https://opentelemetry.io/) and [Prometheus](https://prometheus.io/) exporters.
+