Managed Cluster Hosting for High Traffic Sites | EuroVPS

Managed Clusters

We solve hosting problems for high-traffic sites.

After submitting the form, check your email ;)

Sample Cluster Configuration A

An entry level cluster is ideal for a single high traffic website with traffic up to five million visitors per month. Features a single load balancer, three NGiNX web servers and a dedicated Varnish Cache server HTTP accelerator.
  • 1 x Load Balancer
  • 3 x Webservers
  • 1 x Database Server
  • 1 x File Server
  • 1 x Varnish Server
  • 1 x Private Switch
  • 1 x Backup Server

Sample Cluster Configuration B

A midrange cluster would feature additional webservers, added database and load balancer redundancy, along with redundant file servers, a private switch, and a dedicated backup server. Perfect for if you want additional network and file isolation courtesy of dedicated hardware.
  • 2 x Load Balancers
  • 6 x Web Servers
  • 2 x Database Servers
  • 2 x File Servers
  • 2 x Varnish Servers
  • 1 x Private Switch
  • 1 x Backup Server

Sample Cluster Configuration C

The most advanced cluster configurations feature their own private network uplink (BGP) and a complete 42U managed rack colocation complete with compute, storage, and networking gear in a self-contained pod. If you have tight security requirements and are looking for the characteristics of Private Cloud Hosting, then this is for you.
  • 2 x Load Balancers
  • 8 x Web Servers
  • 3 x Database Servers
  • 2 x File Servers
  • 2 x Varnish Servers
  • 1 x Private Switch
  • 1 x Backup Server
  • 1 x Splunk Server

How to tell if a managed cluster is right for you:

You are obsessed with the safety and reliability of your SQL database

If you are as obsessed with the safety and reliability of your SQL database as we are then you want to know that you can survive a complete physical SQL node failure without any interruption in service. When a failed SQL server is returned to service, it resynchronizes database activity from its partner server and immediately is reinserted into the live database cluster.

You are starting to outgrow even the most powerful dedicated servers

You have probably realised by now that vertically scaling by just upgrading to fast hardware is not enough to continuously guarantee fast page load times. To withstand thousands of concurrent transactions per second, you want a horizontally scalable solution.

You really want to build horizontal scaling into your application

You want to scale tsunami swells of traffic dynamically by adding web and database servers to meet surging demand. But you know you'll need load balancing, horizontal scaling, and a high availability architecture, and it all needs to be built into your application!

You know you need a Devops guy but you can't get one now

Do you need an automated and repeatable way to build, manage, configure environments, and deliver and test code using Chef, Jenkins, or Puppet, but don't have the experience, time or man power to do it?

a few things included in all clusters...

NFS Fault Tolerant Files

Every cluster requires shared storage of some type. Your website & application’s files need a safe home. We like using ZFS because it provides us unlimited file system ‘point in time’ snapshot capabilities.

NGiNX web farm

Redundant webservers read the application’s data files, send queries to the backend Database server(s), and render dynamic pages. Managed Clusters always include at least two or more redundant webservers.

Staging area

Replicate the same configuration of your production environment and test your work before commits. Staging area services are configured identical to Production area services with exception that no provisions will be made for high availability architecture.

Load Balancing

We implement HAproxy clustered load balancing, using hardware heartbeat, to receive and send web traffic to and from your website’s visitors & webservers. Content switching supports dynamic server pool selection, based on any criteria.

Realtime Replication

We are one of the few providers in Europe to offer complete block storage redundancy. If a primary storage system failure, the data stores will immediately “Failover” to the standby SAN without dropping links or timing out.

Varnish Caching

All clusters are setup with Varnish to serve content properly and following your organizations requirements. EuroVPS Priority Support engineers have extensive experience building, maintaining, and configuring Varnish to do exactly what you want.

Support Life Cycle

Fully managed from the physical layer to the software layer.

Consultation

There isn't a single size for a private cluster that can fit all customers. We take the time to find out exactly what your usage case requires to save you money, and time.

Migration

The migration of your web property from any existing hosting to your cluster is managed by us without any interference needed on your part. Emails, blog comments, databases - we migrate it all...

Debug

Once migration is completed, a validation session is organised with your team to confirm proper operation of the website. Any hiccups or problems are debuged along the way by us.

Launch

We're ready for launch! The final step of this procedure, will be to synchronise files and database tables so that when your migration is completed, you have not lost a single post or comment on your websites.

Surf tsunami sized traffic spikes with a managed cluster