1. Home
  2. Training Library
  3. Microsoft Azure
  4. Courses
  5. Storage Solutions for SAP on Azure

Scale-out File Server

Start course
Overview
Difficulty
Intermediate
Duration
18m
Students
12
Ratings
5/5
starstarstarstarstar
Description

Storage Solutions for SAP on Azure builds on storage topics discussed in Design an Azure Infrastructure for SAP Workloads and Design and Build High Availability and Disaster Recovery for SAP Workloads (coming soon). Data storage needs to be fast, responsive, and secure, but above all, continually available.

This course delves into greater detail on previously discussed topics and introduces new, more complex subject matter and its application to SAP workloads to ensure business continuity.

Learning Objectives

  • Understand the various disk types available in Azure
  • Learn how Azure Shared Managed Disks and Storage Spaces Direct can be used for SAP workloads
  • Learn about scale-out file system
  • Understand what disk striping and disk caching are
  • Learn when and how to enable disk write acceleration and how to encrypt disks with SAP workloads

Intended Audience

This course is designed for anyone looking to explore the Azure storage solutions available for SAP workloads.

Prerequisites

To get the most out of this course, you should already have some experience working with SAP and Azure. Before embarking on this course, we recommend you take a look at Design an Azure Infrastructure for SAP Workloads and Design and Build High Availability and Disaster Recovery for SAP Workloads first.

Transcript

Scale-out File Server (SOFS) is a Windows technology first introduced in Server 2012 and is now available in Azure. From a user's point of view, a scale-out file server presents as a single logical server, but behind the scenes, it is a multi-node failover cluster where each node has access to shared storage devices. Scale-out is a bit of a misnomer and would be better described as a mesh architecture because each node can access file content simultaneously. The design is resilient through both redundant storage and multiple nodes. It will perform better than a single server accessing storage due to the increased bandwidth of multiple nodes. 

Storage spaces is a Windows technology that can be likened to software RAID. Drives are combined in ways to provide resiliency, performance, or both, presenting the combination as a single volume that, in turn, can be partitioned into virtual drives. 

While Storage Spaces Direct, also known as S2D, sounds like an evolution of Storage Spaces, it's an entirely different technology, enabling mirroring of locally stored data between servers. It could be described as storage spaces within storage spaces or, more correctly, a virtual SAN. Additional storage can be dynamically added to Storage Space Direct without needing to take the system off-line. Resiliency modes offer characteristics suited to different workloads. Mirror mode is the least space-efficient or most wasteful but offers the best performance, most suitable for database storage. Mirror-accelerated parity is slightly more capacity efficient but at the expense of performance. Dual-parity is the most space-efficient as the number of servers increases but is the least performant.

Combine scale-out file server/shares and Storage Spaces Direct to provide a scalable and continually available SAPMNT file share service. 

While using these two technologies is attractive from a business continuity perspective, there are some caveats to consider. VMs involved in the Storage Spaces Direct cluster have to be within an Availability Set, but they cannot be deployed across different Availability Zones. 

Here are some SAP prerequisites when using Scale-out file shares in Azure. A cluster must have a minimum of two nodes. I guess that's there for completeness, as I'm not sure what you'd call a one node cluster. Each node needs a minimum of two local disks. Three or more cluster nodes are recommended, using three-way mirroring. You can use a two node cluster with two-way mirroring. Azure premium disks are required, and it is recommended that they be managed disks. It's best to format the resultant volumes as Resilient File System. A high-performance network must connect nodes in the Storage Spaces Direct cluster, so make network bandwidth a priority when choosing the VM type.

About the Author
Avatar
Hallam Webber
Software Architect
Students
16360
Courses
30
Learning Paths
5

Hallam is a software architect with over 20 years experience across a wide range of industries. He began his software career as a  Delphi/Interbase disciple but changed his allegiance to Microsoft with its deep and broad ecosystem. While Hallam has designed and crafted custom software utilizing web, mobile and desktop technologies, good quality reliable data is the key to a successful solution. The challenge of quickly turning data into useful information for digestion by humans and machines has led Hallam to specialize in database design and process automation. Showing customers how leverage new technology to change and improve their business processes is one of the key drivers keeping Hallam coming back to the keyboard.