Last updated: 2024-02-01

Introduction

At 8knots, security is a core component of our operations. We are dedicated to maintaining the confidentiality, integrity, and availability of our systems and data. This document provides an overview of our security practices, to demonstrate our commitment to providing a secure and reliable platform for the modern cloud data stack metadata ingestion and analysis.

For further inquiries or discussions related to security, please feel free to reach out to [email protected]

Client data access architecture

We only access client’s usage metadata

8knots does not read, nor does it require access to, any data stored in client-created tables. 8knots only reads usage metadata from tool-specific usage tables or log file as provided by the tools it requires access to. The permissions to the usage matadata is determined during the on-boarding process.

Snowflake usage metadata

8knots reads the usage metadata such as access history, query history, billing/metering history, etc. as made available and standardized by Snowflake. 8knots will read the following information from Snowflake:

Client data extraction

8knots uses a dedicated user/role and warehouse configured on the client’s Snowflake account, with read-only access permissions to usage metadata resources described above, to securely and efficiently copy this data into per-client secured cloud storage files.

8knots uses a combination of AWS, GCP and Snowflake for all of its storage services. The 8knots Web App will periodically read the metadata from the client’s account and store it in flat files on secure S3 or GCS buckets, then ingest that data into 8knot’s internal cloud storage platform where the analysis and services are performed.

Client’s access credentials are stored on 8knot’s GCP cloud services, are encrypted at rest, and cannot be accessed directly by 8knots engineers.