Home > Workload Solutions > SQL Server > Guides > Design Guide—SQL Server 2022 Database Solution with Object Storage on Dell Hardware Stack > Solution introduction
This Dell Validated Design uses Dell’s PowerEdge servers with AMD EPYC 7473X processors, Elastic Cloud Storage and PowerStore array as the underlying solution infrastructure. This design intends to provide solution insights for data engineers, data architects and data scientists who intend to run analytical workloads using Microsoft SQL Server 2022 and object storage.
This design will detail the process of establishing highly available SQL engines for Windows and Linux environments. It also examines the new data analytics and data protection features for SQL Server. This design will demonstrate the implementation of secure and accessible critical infrastructure through this solution setup.
It is vital to select the optimal CPUs for data analytic servers because analytic workloads are regularly CPU intensive. To select the appropriate CPU, it is essential to consider the quantity of cores, their frequency, and the Level 3 cache size.
T-SQL queries for data analytics require quick response times, so AMD 7473X processors were chosen for the SQL Server 2022 database instances. The AMD EPYC 7473X processors have twenty-four cores per socket at 2.8 GHz with a Level 3 cache size of 768 MB which provides enough horsepower for data analytic workloads.
It is crucial to have flexible and scalable S3-compatible object storage and the optimal CPUs for data processing for most analytic workloads.
Dell Elastic Cloud Storage (ECS) is a software-defined, cloud-scale, object storage platform that delivers S3, Atmos, CAS, Swift, NFSv3, and HDFS storage services on a single, modern platform. It provides simple RESTful API access for storage services. Dell ECS provides significant value for organizations seeking a platform that supports rapid data growth. The advantages and features of Dell ECS include:
Cloud Scale Storage
Flexible deployment
Total Cost of Ownership reduction
Microsoft SQL Server is widely used across all industries, and these data sources are commonly mission critical. The ability to backup and restore SQL Server databases is crucial. The ability of SQL Server 2022 to backup and restore to S3-compatible object storage provides additional flexibility through cloud connectivity. To use this feature, T-SQL provides the TO URL syntax for backup and FROM URL syntax for restore.
SQL Server 2022 PolyBase makes data virtualization possible for data scientists using T-SQL for analytic workloads, by querying data directly from other sources such as Oracle, Teradata, Hadoop cluster, and S3-compatible object storage without separately installing client connection software. PolyBase allows T-SQL queries to join data from external sources with relational tables in an instance of SQL Server. The T-SQL OPENROWSET and EXTERNAL TABLE syntaxes are useful for querying data in S3-compatible storage.
Data virtualization is a broad term that describes a data management approach. It allows an application to retrieve and manipulate data without requiring the data’s technical details, such as the data’s physical location and source format. Data virtualization involves abstracting various sources through a single data access layer. Organizations are adopting tools and software to integrate different types of data virtually. This data integration enables both data mining and analytics, and it is critical for predictive analytics tools for use of machine learning (ML) and artificial intelligence (AI).
Dell solutions engineers created two physical architecture setups, using VMware virtualization and Red Hat OpenShift with SQL Server 2022 to test analytic workloads using T-SQL. These architectures use Dell ECS for external data and PowerStore for the data found in the local SQL Server 2022 instance.
The scope of this Dell Validated Design paper is to offer a database solution for data engineers, data scientists, and architects. This design guide is intended for enterprises who will run analytic workloads on SQL Server 2022 with object storage and use Dell PowerEdge servers with AMD EPYC 7473X processors, Dell ECS, and a Dell PowerStore storage array as the underlying infrastructure.