As the digital landscape expands, businesses need to manage their VMware SD-WAN™ and VMware SASE™ services efficiently and securely. A session from VMware Explore Las Vegas called “Gain Actionable Insights from Thousands of VMware SD-WAN and SASE Deployments” explored this in detail.
To dive deeper into these insights, watch the video below. As the data landscape continues to evolve, understanding these mechanisms and tools is key to staying ahead of the curve.
The session was presented by members of the business intelligence team in the SASE Center of Excellence. They focus on understanding and collecting data to help VMware by Broadcom understand network performance and customer deployments. For example, sales teams can use the data to process renewals and look at edge counts, bandwidth configuration and usage. Support teams can notify customers when they are breaking best practices or have potential to hit a bug. Product teams are able to measure which features customers are using.
Using real data to measure network performance
In networking, one question stands out: How is my network performing? While you can check each site within the network, this approach becomes impractical for large networks. This session demonstrates how to address this challenge, showcasing the power of APIs and the insights they provide through data mapping. By the end, you will gain a deep appreciation for the immense value of VMware SD-WAN and VMware SASE data and the actionable insights they offer.
To put things into perspective, consider the scale of our analysis. Each year, we analyze over 4.5 trillion data points from approximately 18,000 customers and hundreds of thousands of edges. The sheer volume of data poses several challenges. How can we effectively process such a vast amount of data? How do we analyze it and identify potential issues? How can we ensure that our solutions cater to the diverse needs of different users? And perhaps most crucially, how do we transform trillions of data points into valuable and actionable insights?
One way we tackle these challenges is by using customer data. We can extract valuable insights by understanding where our customers are located, their specific industries, size, and revenue. These insights not only inform decision-making processes but also provide a comprehensive understanding of various industries.
We can also dissect customer networks to see how well the network is performing. It is essential to assess the network as a whole, especially when troubleshooting performance issues. We can examine metrics including CPU and memory consumption, link and overlay utilization, and queue drops to understand the underlying causes.
We collect 288 data points daily across multiple key metrics. These statistics are then aggregated into monthly data, allowing us to support a wide range of use cases. Whether detailed information is required by engineers or high-level summaries are needed by executives, our data caters to all users’ needs.
The role of APIs in collecting and analyzing data
At the core of the session was the importance of APIs (Application Programming Interfaces) and the role they play in collecting and analyzing data. VMware’s data pipeline begins with the orchestrator APIs that collect data. This data is then retrieved by a Python-based collector and dispatched to a caching layer where it is validated and stored as JSON files in an S3 bucket. Following this, an ETL (Extract, Transform, Load) pipeline processes the data, effectively integrating it with customer insights within a PostgreSQL database, chosen for its UUID support and general familiarity.
The resultant data has multiple applications, notably in what the CoE calls low-touch Customer Success programs that aim to improve the customer’s experience and success with VMware SASE and VMware SD-WAN. These initiatives have a customer-centric focus, employing data-driven strategies to improve user experience.
“Low touch” means that no individual work is required for each customer. Similar to a Netflix recommendation email, the programs draw customers back to the platform to enjoy it more. Communications to partners and customers could include maintenance emails, recommended versions, and informative knowledge base articles that proactively communicate updates, bug fixes, and best practices to users.
Data visualization and analysis
Beyond this, the data is also leveraged for visualization and analysis. Tools like Power BI are employed to create rich, detailed insights, uncover trends, and generate comprehensive reports. The session also hinted at ongoing advancements in AIOps, particularly for trend analysis and anomaly detection.
This approach to network analysis involves harnessing the power of APIs, leveraging customer data, and engaging in thorough assessment of each network component. The insights and solutions that VMware by Broadcom provides cater to a diverse range of users and industries. With expertise and robust data analytics, we can help you optimize your network’s performance and drive informed decision-making.
