# USE CASES


High availability

DRP

fault tolerance

ERP hosting

critical application

Multi-datacentre infrastructure

3-tier architecture

MySQL replication

Load Balancing IP

dedicated server

Backup server

So you Start

OVH.ie


Implementing a HA infrastructure with a disaster recovery plan (DRP) for critical application hosting



A firm of 70 expert accountants is one of the customers of an IT services company. The company originally chose to host the firm's ERP (business process management software) on a Cloud VPS, which provides high SLA to guarantee maximum availability for this critical application. As the number of ERP users increased, the capacity of the VPS Cloud became insufficient. It reached a point where the firm needed to migrate to a dedicated server, but for this the firm had to think about how to mitigate potential hardware failures. In the end, it deployed a high availability infrastructure based on two identical servers, both of which connect to a backup server that secures the data as part of a disaster recovery plan.

This use case was inspired by the infrastructure that OVH France customer EURL Barme built for its Paris-based customer Sextant Expertise, and for which EURL Barme won the 2014 OVH Infrastructure Award in the Company Infrastructures (Outsourcing) category.




Key information


  • Hosting an accountancy firm's ERP (business critical application)
  • 70 daily users
  • 2 identical dedicated servers, one hosted in SBG and the other in RBX
  • Backup on a So you Start server
  • VPS Cloud for HA monitoring

Specifications


  • Internet access to the ERP secured by an SSL certificate
  • High availability for the firm's critical application
  • Data security: the data is sensitive - it includes financial information - and mustn't be lost or leaked




Solution deployed at OVH





When we have to host a critical application, our first reaction is to use two servers: an active server (master), and a passive server (slave) to take over if the master fails. This is an entirely plausible solution for a disaster recovery plan (DRP) built around two metrics: RTO (the maximum allowable interruption) and RPO (the maximum targeted period in which data might be lost). However, for a slightly higher budget and a fairly small infrastructure, it's possible - and often beneficial - to build a high availability infrastructure with two active servers. This kind of infrastructure has two big advantages. One: you can transfer some of the load on to the second server; two: if one server fails, the application and service will continue to run, i.e. there will be NO downtime. The VPS that originally hosted the infrastructure can be recycled and used as a monitoring machine.

1 – Distributing the load

2 – An infrastructure distributed over two active servers that are hosted in 2 separate datacentres

3 – Replicating the database between the two servers

4 - Backup on a So you Start server

5 - Monitoring



For more info


If the number of ERP users were to grow quickly, you would be able to scale up the infrastructure by allocating the database front-ends (web and application layers) to several servers. You can then increase the number of database servers to spread the load over a larger number of servers. You can also reduce backup costs by saving backups on the RunAbove object storage platform. The goal: to only pay for the amount of storage required for the quantity of data you have, and not for hard drive resources or a server with underused CPU. Moreover, RunAbove Storage automatically replicates your data 3 times so that your data backups are extra secure.



Infrastructure components




1 Load Balancing IP

9.99 ex. VAT/month

Buy now


Front-end servers

2 Infrastructure servers
Model EG-32

86.99 ex. VAT/month

Buy now


Private connection between servers

vRack
Service included with OVH.com Infrastructure range servers


Backup servers

So you Start DISK+ servers
Model BK-8T

52.99 ex. VAT/month

Buy now


Monitoring

1 VPS Cloud 1

9.99 ex. VAT/month

Buy now