Aviz ONE-Data Lake details integration with AWS S3 for ONES 2.1 metrics

Aviz ONE-Data Lake, introduced as part of ONES 2.1, adds a documented integration path to store network telemetry in an AWS S3 bucket. The update matters to enterprise security and operations teams that need centralized collection, controlled access, and workable analytics across multi-vendor network environments.

Research Overview

The blog describes ONES 2.1’s ONE Data Lake as a cloud repository for network telemetry gathered from on-premises network environments. It frames the S3 integration as a way to stream operational metrics into a designated S3 location for later analysis.

It also reiterates that a data lake architecture typically uses scalable cloud storage for structured, semi-structured, and unstructured data. The article positions AWS S3 as the storage layer used for this telemetry workflow.

Key Findings

ONE-Data Lake supports migrating network data into cloud storage and includes metrics that cover the network control plane, data plane, system, platform, and traffic. The blog states that ONES Cloud stores the metrics previously used in ONES.

For the S3 integration, the article outlines configuration elements that include an IAM role identifier, an AWS region, and an S3 bucket name. It also adds that an external ID is optional for cross-account access.

Technical Breakdown

The S3 integration is described as an ONES Cloud setup process where S3 instances are configured so ONES can push metrics to a designated cloud endpoint. The blog’s required parameters are the ARN role, region, and bucket name, plus an optional external ID.

After integration, the blog states that the created cloud instance in ONES can be updated, paused or resumed, and deleted. It also describes metric selection as user-defined, with administrators choosing which categories and specific metrics to stream.

Metric selection and supported telemetry

The blog says ONES 2.1 supports metrics grouped under Traffic Statistics, ASIC Capacity, Device Health, and Inventory. It states administrators can choose and deselect metrics from the available list within these categories.

It further states that ONE-Data Lake’s collection is not limited to a single vendor or network operating system. It specifies device coverage including Cisco NX-OS, Arista AOS, SONiC, and non-SONiC platforms.

Streaming method across network operating systems

The article states that streaming uses gNMI for SONiC-supported devices and SNMP for other vendors’ operating systems. It includes a reference to ONES inventory showing multiple vendor devices streaming data.

S3 analytical capabilities described in the blog

The blog lists multiple ways to analyze data stored in S3. It describes AWS Athena for serverless SQL queries on S3 data and gives examples such as querying log files and data formats including CSV, JSON, and Parquet without setting up a database.

It also describes AWS Glue as a managed ETL service for preparing and transforming data, and AWS SageMaker for building, training, and deploying machine learning models using datasets stored in S3. The blog adds that third-party tools can integrate with S3 and cites examples including Databricks, Snowflake, and Domo.

Operational Impact

The blog’s S3 integration steps emphasize operational management within ONES, including the ability to update integration settings and to pause or resume metric uploads to the cloud. It also describes deletion of the integration created within ONES.

For storage, the article describes S3 features relevant to durability, security, and lifecycle management, including encryption at rest and in transit, IAM-based access controls, lifecycle policies, versioning, and replication. It also lists AWS storage classes such as S3 Standard, S3 Intelligent-Tiering, S3 Standard-IA, S3 One Zone-IA, and S3 Glacier as part of the cost model discussion.

Conclusion

The blog explains how ONES 2.1’s Aviz ONE-Data Lake integrates with AWS S3 by using an ONES Cloud configuration workflow, user-selected telemetry metrics, and multi-vendor streaming methods, followed by analysis options such as Athena, Glue, and SageMaker. This “Blog Signals brief” is a fact-based summary of the vendor blog.