---
title: "Best and Worst High Availability Use Cases: A SafeKit Implementation Guide"
canonical: "https://safekit.eviden.com/use-cases/best-worst-use-cases-high-availability/"
description: "Identify the best and worst use cases for high availability. Learn where SafeKit excels in software-defined clustering for Windows and Linux, and discover when traditional shared-storage or hardware-based HA might fail your business."
category: "use-cases"
lang: "en"
topics: "Best use cases of SafeKit high availability, Worst use cases of SafeKit high availability, SafeKit High Availability Limitations"
---

# Best and Worst High Availability Use Cases: A SafeKit Implementation Guide

##  Best use cases of SafeKit high availability 

### OEM Software

A software publisher uses [SafeKit as an OEM software](<https://safekit.eviden.com/use-cases/application-clustering-software/>) for high availability of its application

![Application Clustering Software for a Software Publisher](https://safekit.eviden.com/wp-content/uploads/2023/02/application-clustering-software-3.png "w:300")

### Distributed Enterprise

A distributed enterprise deploys [SafeKit in many branches](<https://safekit.eviden.com/use-cases/business-continuity-software/>) without specific IT skills

![High Availability Software in a Distributed Enterprise](https://safekit.eviden.com/wp-content/uploads/2023/02/distributed-enterprise-1.png "w:300")

### Remote Sites

[SafeKit is deployed in two remote sites](<https://safekit.eviden.com/best-practises/high-availability-disaster-recovery-hadr/>) without the need for replicated bays of disks through a SAN.

![Business Continuity and Disaster Recovery without a replicated SAN](https://safekit.eviden.com/wp-content/uploads/2023/02/remote-datacenters.png "w:300")

##  Worst use cases of SafeKit high availability 

## SafeKit High Availability Limitations {#safekit-ha-limitations}

Limit / Requirement | Reason & Impact | Recommended Alternative  
---|---|---  
Replication size: A few TBs |  Resynchronization time after failure ([step 3](https://safekit.eviden.com/architectures/file-replication-byte-level-with-failover-mirror-cluster/#safekit-mirror-cluster)): <br>• 1 Gb/s network ≈ 3 hours / TB<br>• 10 Gb/s network ≈ 1 hour / TB (depending on disk write performances)  |  • Use [external shared storage](https://safekit.eviden.com/best-practises/san-vs-nas-shared-storage-for-a-failover-cluster/) (more expensive, more complex).   
File count: < 1,000,000 files |  • Resynchronization time performance after a failure ([step 3](https://safekit.eviden.com/architectures/file-replication-byte-level-with-failover-mirror-cluster/#safekit-mirror-cluster)).<br>• Time to check each file between both nodes.  |  • Put the many files to replicate in a virtual hard disk / virtual machine.<br>• Only the files representing the virtual hard disk / virtual machine will be replicated and resynchronized in this case.   
Failover capacity: ≤ 32 VMs |  • Each VM runs in an independent mirror module.<br>• Maximum of 32 mirror modules running on the same cluster.  |  • Use an external shared storage and another VM clustering solution (more expensive, more complex).   
Network architecture: LAN/VLAN network between remote sites |  • Automatic failover of the [virtual IP address](https://safekit.eviden.com/best-practises/how-a-virtual-ip-address-works/) with 2 nodes in the same subnet.<br>• Good bandwidth for resynchronization ([step 3](https://safekit.eviden.com/architectures/file-replication-byte-level-with-failover-mirror-cluster/#safekit-mirror-cluster)) and good latency for [synchronous replication](https://safekit.eviden.com/best-practises/synchronous-replication-vs-asynchronous-replication/) (typically a round-trip of less than 2ms).  |  • Use a [load balancer for the virtual IP address](https://safekit.eviden.com/best-practises/how-a-virtual-ip-address-works/) if the 2 nodes are in 2 subnets (supported by SafeKit, especially in the cloud).<br>• Use backup solutions with asynchronous replication for high latency network.