Building a Resilient ESB Cloud using Availability Zones

When Pentagull began evaluating cloud-computing providers towards the end of 2016 in readiness for launching our ESB cloud, one of our key design objectives was resilience. This may seem obvious, as every cloud provider claims to have extremely high levels of resilience – but we weren’t interested in uptime guarantees and service level agreements – we wanted hard facts about the design of the infrastructure and what features it would provide to back up those uptime promises in real-world failure scenarios. After an extensive evaluation period we eventually chose AWS as our provider – and one of the key factors in that decision was a feature called Availability Zones.

Most cloud providers offer multiple geographically dispersed regions in which they provide cloud services, and AWS is no exception in this regard. However, AWS goes one step further by offering multiple data centres within each region – known as Availability Zones. These data centres are still sufficiently isolated as to protect against the majority of infrastructure issues, but are close enough to offer extremely low latency, high bandwidth links. This makes a huge difference when providing load balanced, redundant services and clustered databases: It is a lot harder to perform effective real-time replication across geographic regions where significant latency may be present. By contrast, the dedicated low latency links between availability zones are perfect for distributed redundancy.

In addition, many of our customers require that their data remains within the UK at all times, so choosing a cloud partner that relies on using alternate locations within the EU for disaster recovery was not an option for us. AWS currently provides a total of 3 UK-based locations, and we make use of all of these in our DR planning. AWS are no longer the only provider to offer Availability Zones, however we believe the fact that AWS was designed and built from the ground up with Availability Zones at its core, rather than as an add-on, is a key advantage when it comes to overall stability.

Our choice of AWS as our cloud services partner will allow us to deliver industry-leading performance and reliability to our ESB cloud customers for many years to come. If you would like to learn more about our ESB cloud offerings, please contact us.

Other news stories

Password Advice
Password Advice

We suggest to both our employees and customers that employing a strong password, with a mix of capitalised letters, numbers and special characters is the most effective way of reducing the chance of being compromised by unauthorised access. However, we don’t force regular password changes for several reasons.

ESB 22.01 Released
ESB 22.01 Released

Pentagull is pleased to announce the release of version 22.01 of the ESB platform. This release continues the recent trend of incremental improvements to reliability and usability as well as providing some feature enhancements.

Can you help to bring some extra joy to children this Christmas?
Can you help to bring some extra joy to children this Christmas?

Could you help spread a little extra festive cheer to care leavers in Blackpool?

Apache Log4j Vulnerability all clear for ESB
Apache Log4j Vulnerability all clear for ESB

A quick update for our customers regarding the recently discovered Apache Log4j vulnerability.

Merry christmas from everyone at Pentagull!
Merry christmas from everyone at Pentagull!

Merry Christmas to all our customers from everyone here at Pentagull!

ESB 21.10 Released
ESB 21.10 Released

Pentagull is pleased to announce the release of version 21.10 of the ESB platform. This release includes a mix of new features as well as refinements to usability and reliability.