Get a closer look at our multihomed network management

Many dedicated server providers like to keep details about their network operations and internal processes behind closed doors. We like to do the opposite and try to share as much information with you as possible, why?

Because we believe that openness, fairness, and confidence in what we do and share is the best way to build lasting relationships, be it personal, or business. In this post we will share with you the basics of how we keep our network quality to the max and how we resolve any problems if they occur with minimum delay.

The very nature of our business is delivering content (data) to your end users with the highest speed and 100% quality. The volume of data running through our network is constantly growing and we have recently surpassed 1Tbps of daily data flows worldwide!

The type of content varies as we accommodate a wide range of application types on our servers ranging from Streaming services, VOD,  Ad delivery infrastructures to Gaming and many more. The traffic patterns throughout these industries tend to be irregular and difficult to predict and that is when our unmetered bandwidth dedicated server with a 10GE uplink comes in handy as it offers enough reserved capacity for any unexpected peaks.

bandwidth-peak-70gbps

However, with the volume of customer traffic constantly growing, the need rises to upgrade our IP upstream capacity as well as the network backbone hardware components.

As we like to have a safety net of at least double the daily network workload, we‘re increasing our IP upstream capacity to 3Tbps, as the demand for bandwidth is growing worldwide, further upgrades can be expected in the upcoming months. The network capacity is ensured by 100GE port connections to multiple top tier transit providers.

To allow seamless upgrading of upstream capacity as the network grows, we regularly upgrade the core routers across our EU & US datacenter locations. During the last upgrade, we’ve sent one of our network architecture gurus on a small “road trip” across the US DC locations for 14 days!

network_architect_trip

Here is a selfie he sent us after landing in Los Angeles – obviously jetlagged from the long flight from London, but ready to supervise the necessary network hardware changes!
All in all, our current outgoing port capacity after installing new core routers, is more than 10Tbps. Therefore plenty of headroom for future upgrades.

Flexible routing with in-house built monitoring system

As mentioned before, excess network capacity from multiple carriers is imperative for accommodating any unexpected traffic peaks, it is also crucial for offloading large-volume DDOS attacks. However sufficient amount of network capacity is not enough for a world scale network to run smoothly and without any packets being lost along the way.

What is also essential is a properly set up live monitoring system that can detect issues in network paths and provide long-term analysis of the upstream path quality. There are a few third-party tools built for this purpose, but none of them met our requirements in full.

That is why we decided to build our own custom network monitoring tool in-house.

This tool is used to monitor routes to problematic destinations in real-time. If you report an issue with one of your end users, the details are automatically added to the system which immediately starts analyzing all available network paths (through multiple carrier backbones). If the current path is congested, we automatically switch to the best performing, unsaturated route to avoid any packet loss.

Real-life scenario:

Let’s say you’re are a streticke-support-systemaming service provider and you receive a call from one of your subscribers pointing out poor quality of the streams he is receiving.

All you need to do is raise a ticket through our client portal with the necessary information:

1) Specify which server are you distributing the content from.

2) Enter the IP address/es of the problematic destination.

Information is then instantly added to our monitoring system for analysis. Once a suitable alternative network path is determined, we reroute the traffic to achieve best network performance.

This flexibility is possible as we have multiple upstream providers and quality partnerships with them.

 

 

All mentioned features and services are standard across all of our bandwidth plans. And if you’re currently not our customer and have servers with a different provider, but regular phone calls from unhappy customers are causing wrinkles on your forehead. Try our 7-day trial with unshared and unmetered 10Gbps bandwidth and test the performance for yourself. Be sure to describe your requirements and concerns you would like to solve in the notes so we can start working on a perfect solution for you right away 🙂