White Papers

Introduction
4 Dell EMC PowerVault ME4024 8,000 Mailbox Exchange 2016 Resiliency Storage Solution using 10K drives | 3917-ESRP
1 Introduction
This document describes the Dell EMC PowerVault™ ME4024 solution for Microsoft
®
Exchange Server,
based on the Microsoft Exchange Solution Reviewed Program (ESRP) Storage program.
This document details the performance characteristics of ME4024 arrays in a fully hardware-redundant
configuration running Microsoft Exchange Server 2016. The solution includes 8,000 typical mailbox users
running on two 2U ME4024 arrays with 24 10K (10,000 rpm) 1.2 TB hard drives. The results show the
ME4024 provides more than enough I/O performance while keeping latencies low.
The ESRP Storage program was developed by the Microsoft Corporation to provide a common storage
testing framework for vendors to provide information on storage solutions for Microsoft Exchange Server
software. For more details on the Microsoft ESRP Storage program, refer to
https://technet.microsoft.com/enus/office/dn756396.aspx.
1.1 Simulated environment
The solution presented in this document is designed to simulate a moderate number of mailboxes hosted on
highly redundant hardware. Application-level redundancy is augmented with redundant storage to create a
highly available and fault-tolerant solution.
The mailbox resiliency features of Exchange Server 2016 greatly enhance the availability of Exchange
Server, while also improving I/O performance. The solution presented is a mailbox resiliency solution utilizing
one database availability group (DAG) and two copies of every database. The tested environment simulates
all users in this DAG running on a single ME4024 array. This is to simulate one half of the solution running the
storage load in a failure scenario.
The number of users simulated was 8,000 concurrent users with 2,000 users per server. The mailbox size
was 1 GB per user. Each server was configured with five databases, with one local copy simulating replication
to a second copy. The scenario simulates a full-redundant solution from both a hardware and software
perspective.
The replication mechanism is the native Exchange 2016 DAG database replication engine. This is an efficient
and reliable replication mechanism and is the recommended method for providing highly available and
redundant Exchange solutions.
1.2 Solution description
The testing environment consisted of one ME4024 array with redundant controllers, each with two 10Gb
iSCSI ports. All ports on the ME4024 are active with Microsoft MPIO round-robin with ALUA, ensuring best-
path access. The Ethernet switches were configured in redundant networks to provide network fault tolerance.
The storage utilized the internal 24-drive-bay enclosure on the ME4024. All drives used were 2.5-inch 10K
rpm 12Gb SAS hard drives. For data protection, the drives were configured with ME4 Series ADAPT RAID
technology. This provides fault tolerance with distributed sparing and the ability to expand the storage with
zero downtime. Because this is a redundant solution, databases and logs are stored together on the same
volumes.