HTAP and Azure SQL for cost savings in energy management

In this article, I would like to expose how a hybrid transactional and analytic processing (HTAP) architecture solution is an example of how an Azure cloud infrastructure can help energy providers that need to combine device data with analytics to manage smart grids in saving their costs. Typically, in order to process transactions, most current systems use low-latency, high-volume operational workloads.

In this particular analytics case, the workloads are higher latency and lower volume, so HTAP architectures provide an optimal solution for both types of workloads because, by using in-memory databases, HTAP consolidates technologies to optimize queries on large volumes of data.

To implement this case, we will rely on Azure SQL Database as the core of the HTAP solution. The approach is to partition the data into horizontally distributed databases or partitions. Additionally, we will employ the following solutions to implement the model:

  • Azure Event Hubs for data ingestion.
  • Azure Stream Analytics for data processing.
  • Azure Functions for partitioning.
  • Azure Blob Storage for event storage.

A solution for the analysis and management of smart grids

What is the process like?

1. Event Hubs is responsible for ingesting telemetry from local facilities (metering).

2. Blob Storage will capture the data from Event Hubs and store it for future analysis.

3. Stream Analytics will process the data. In the active path of the solution, Azure Cosmos DB will be in charge of querying the data from the previous two months. It is important to note that Azure Cosmos DB guarantees response times of less than 10 milliseconds so it does not substantially increase the operation time.

4.  If errors occur during data processing or storage, the system records them in Azure Table Storage.

5. Azure Functions will use the SQL Database elastic database client library to archive the data. This process will partition the data to optimize insert operations. The solution partitions the data by distributing the data horizontally across multiple Azure SQL databases. Each database uses a partitioned clustered column index to compress the tables. Response times on this idle path are typically less than one second.

6.  An Azure Databricks cluster reprocesses data from Blob Storage. Specifically, Azure Databricks deserializes the Avro files and sends the data to Event Hubs for optional analysis.

Together, these services provide an HTAP solution that:

  • It produces a significant cost reduction in energy management by providing fast access to conclusions about archived data. With this solution, latencies in the sporadic access level path drop from hours to less than seconds.
  • Simplifies archiving by automatically adding data to long-term storage.
  • Maximizes scalability by creating data partitions and using an elastic database.

What do you think about it? If you have any questions or think we can help your organization, don't hesitate to contact us. We will be happy to help you😊