TechPrufer
Get started
Pinecone
Pinecone logoPinecone · Feature trail

Bring Your Own Cloud (BYOC)

4 posts · Jul 2023 – Sep 2026

What this trail covers

Pinecone serverless is now generally available on AWS, GCP, and Azure with a Bring Your Own Cloud (BYOC) offering. This allows customers to deploy a privately managed Pinecone region within their own cloud account, ensuring data sovereignty and control. The BYOC offering includes a zero-access operating model, self-serve deployment, and support for AWS, GCP, and Azure. It complements the existing serverless offering by providing enhanced security and compliance for enterprises and regulated industries.

All Pinecone trails
  1. Pinecone BYOC: Trusted AI Knowledge in the Customer CloudThis post announces the general availability of Pinecone Bring Your Own Cloud (BYOC) on AWS, GCP, and Azure. The BYOC model allows the data plane to run within the customer's cloud account, ensuring sensitive data remains within their controlled boundary. It details a zero-access model where Pinecone manages the service via outbound calls from the control plane, without requiring SSH, VPN, or inbound network access. This architecture maintains the same APIs and interfaces as the managed service, offering flexibility for different security and operational requirements.
  2. Pinecone BYOC: Pinecone in your AWS, GCP, or Azure account, no vendor accessIntroduced Pinecone Bring Your Own Cloud (BYOC) for Enterprise users in public preview. BYOC enables the Pinecone data plane to run within a customer's AWS, GCP, or Azure VPC with a zero-access operating model. This means Pinecone does not require SSH, VPN, or inbound network access to the customer's cloud account, VPC, or cluster. Control plane operations are pulled outbound via TLS, and data plane requests are served directly from within the customer's VPC. Operational metrics and traces are sent to Pinecone for monitoring, while vectors, metadata, and request payloads remain within the customer's cloud account. Connectivity options include public endpoints or private connectivity via AWS PrivateLink, GCP Private Service Connect, or Azure Private Link. Deployment is facilitated by a self-serve Pulumi kit. Currently supports single-namespace Dedicated Read Nodes (DRN) indexes.
  3. Bringing the leading vector database to your cloudThis post announces the early access of Pinecone's Bring Your Own Cloud (BYOC) offering for AWS. It details how BYOC allows customers to deploy a privately managed Pinecone region within their AWS account, ensuring data sovereignty and control. The architecture is described as a split model with a customer-hosted Data Plane and a Pinecone-hosted Control Plane. Key benefits highlighted include enhanced security, compliance, enterprise-grade access controls, private multi-tenant regions, deployment in any AWS region, and potential cost savings by leveraging existing AWS discounts. The post also mentions the use of RBAC with IAM permissions and boundary policies for security, and a customer-controlled VPN for maintenance access.
  4. Pinecone serverless on AWS is generally availableThis post announces the general availability of Pinecone serverless on AWS and introduces Private Endpoints for AWS PrivateLink in public preview. It details the architectural innovations in the write and read paths of Pinecone serverless, emphasizing the separation of reads, writes, and storage, vector clustering on object storage, and innovative indexing and retrieval algorithms. The post also highlights the benefits of Private Endpoints for enhanced security and data traffic protection over Amazon's private network.
  5. Run Pinecone on AzurePinecone is now available on Azure, starting with early access in the 'eastus-azure' region for Standard and Enterprise users. This expansion aims to provide performance at scale by having Pinecone closer to Azure-based data, applications, and models, leading to lower end-to-end latencies. It also simplifies procurement and helps customers meet enterprise readiness requirements by running on their preferred cloud provider. The initial offering supports p1 and s1 pods, with plans for Azure marketplace availability later in the year.