UK Cloud Infrastructure Upgrades

Were incredibly excited to announce that marilyn and i are underway with a big overhaul of our UK based infrastructure that is rolled out over the following few months.

For our continued effort in order to at the forefront of our industry, you'll find spent another 18 months planning for a large-scale upgrade to all your hosting platform, introducing a core concentration on cloud reliability, increased performance and scalability. 

The use of the upgrade are generally outlined throughout this page. 

Huge-Scale Cloud Infrastructure to switch our Dedicated Hardware

UK Cloud Infrastructure Upgrades

Dedicated Cloud Infrastructure
At this time, all of our UK based accounts are hosted on physical hardware in London, UK. Whilst this offers fantastic power and also gratifaction, we do realize that dedicated hardware lacks in other regions, for example scalability, redundancy and reliability.

Our new platform has become planned and built from ground up, to present a huge number of benefits over our existing hosting platform.

More Servers, with Better Distribution of Accounts
Improved Contention
Our current hosting environment consists of enormous specification dedicated hardware, because both versions hosts many accounts per physical server.

  Contained in the planned upgrades, we are creating a much bigger number of ‘smaller'machines, because both versions will host an inferior number of accounts (approximately 50% less accounts per-node than our current platform).

It is very important to point out here that her ‘larger'server certainly doesn't mean more performance (this is a kind of misconception). Smaller machines, with less contention / accounts can deliver significantly higher performance, good results . several added advantages from an administrative point of view.

The performance of sites however, is there to greater degree, driven via the architecture and components of such nodes as an alternative to the physical ‘size '. The alterations on these areas are outlined below.

100% high-availability NVMe storage cluster
Pure NVMe Storage Cluster

UK Cloud Infrastructure Upgrades

Talking of performance – we decided to enhance our private cloud environment for the huge variety of pure Enterprise Grade NVMe storage, as an alternative to SSD drives!

In terms of per-disk performance, the visible difference is phenomenal. This takes our storage performance to a very new level, with faster sequential read / write speeds and a brilliant uplift in IOPS.

The drives you'll find chosen to make use of for the environment are U2 formfactor, hot swappable devices, that can be added / removed in the cluster with 0 interruption to service (see more information on our ‘data replication'and clustering below).

But not just shall we be held introducing adore the look uplift in performance, but we will also be introducing triple-replication to our infrastructure for improved redundancy and failover.

Enterprise-Grade CPU Upgrades
AMD EPYC™ CPU
Our revised cloud infrastructure is entirely powered via the incredible AMD EPYC™ Enterprise Grade CPUs throughout, boasting an unrivalled degree of power and also gratifaction per core.

The popular CPUs will introduce a 300% uplift in memory bandwidth, allowing much more quickly RAM handling overall, along with a heavily increased clock-speed per CPU core.

The popular CPUs will assist you to achieve greater per-core performance for use on your websites and applications.

High-Availability Infrastructure and Data-Replication
Data Replication, Clustering and Failover
We certainly have built our new cloud platform for being incredibly robust with self-healing capabilities, auto-failover and triple-replication of your respective data.

Our infrastructure has become designed and battle-tested to face up to power outages, hardware failures and routine hardware upgrades with no need for interruption to all your services.

Our new hyper-converged storage platform (based on 100% NVMe storage) has intelligent, self-healing capabilities allowing us to seamlessly replace physical disks without the need for downtime. The equivalent storage platform will definitely withstand entire node failures, again, without impact to service.

  Our storage option is not based on a ‘SAN'type environment, but alternatively a hyper-converged storage cluster between our hypervisor / compute nodes. This allows us to infinitely scale our storage platform as our demand as well as grow, without disruption in the virtual machines running on the cluster.

Our physical servers are interlinked with 80Gbps of connectivity, yet are physically isolated from each other well within different racks of the details center.

Which means that any time an entire rack outage, the virtual machines which hosts your websites will failover immediately to among the remaining servers / racks without disruption, or degradation to service or performance.

Huge network capacity uplifts
Increased Network Capacity
At this time, each of our dedicated servers runs a single 1Gbps public port, which is used to serve traffic / general bandwidth, and backups.

  The popular cloud platform boasts a huge 20Gbps public connectivity per-node, in addition to 80Gbps interconnectivity between hypervisors.

  Also, you'll find introduced a direct 100Gbps link from London in the Data Center in Frankfurt, which is used for the backups and snapshots.

  This huge uplift in network connectivity provides faster site loading times, massively increased bandwidth capacity and drastically improved backup performance.

Stable, Powerful and Secure OS, devised for web hosting service
Cloudlinux 8
All machines in your private cloud will undoubtedly be provisioned using the next stable and secure relieve Cloudlinux 8 that should run the 4.18 Cloudlinux / LVE kernel.

Cloudlinux is the very best that marilyn and i manipulate and relied on many different years, which introduces very voluminous benefits to any shared hosting platform for example LVE, CageFS and more.

Cloudlinux 8 introduces several benefits over the last version to help promote the capabilities of such tools and facilities it comes down bundled with.

Improved DNS Latency, Performance and Security
Improved DNS Clustering
For our upgrade, we are relocating our UK based nameservers from Coventry / Maidenhead, to London to enhance latency and performance.

To be sure a seamless migration process, we are still make it possible for DNS zones are synchronised to the actual cluster, allowing time for that required changes for being designed to any custom nameservers or domains.

More info on the migration process will undoubtedly be provided over the following few weeks. During this period, no changes are required.

Faster, more reliable backups
Improved Backup Performance
Introducing the mix of our 20Gbps public links and our direct connectivity with Frankfurt, we are now able to increase our daily backup performance dramatically.

Also, for the reason that each server will host less data and smaller quantities of accounts, we estimate to relieve our backup window by at any rate 40-60%.

In addition, but we are introducing weekly backups of this virtual machines which will power your websites, so that marilyn and i employ a full recovery point of the full machine for emergency scenarios. full-width

4 Comments

Previous Post Next Post