Subscribe

Is your cloud solution bullet-proof?

Services will go down at some point, but proactive planning can make the world of difference.

Kevin Hall
By Kevin Hall, National sales manager, Elingo
Johannesburg, 22 Jul 2016

In a world where uptime guarantee is the order of the day, and the clients and consumers demand 24/7 availability from the vendor, how does one emerge as an always-on service and meet the needs of a growing user base?

The reality is that services will go down at some stage, and the level of preparation and preventative planning will make all the difference to user experience.

Planning for emergencies, and truly testing the worst-case scenarios, is the most effective way to go about things. If the client trusts the vendor, then the vendor needs to make sure that trust is met with the utmost effort to ensure reliability.

The architecture in cloud offerings lends itself to many solutions and options to provide the end-user with the peace of mind that the technology can - and will - provide service all the time, every time.

* Make sure that if the client wants to access the vendor's service and solution, it is easy and user friendly

Vendors want to have the client using what they have as much as possible to accomplish daily tasks. The cloud provider has a month-to-month arrangement, so it is in the interest of the vendor to make sure the end-user is well trained, and understands the value the technology offers.

If every user questions the reliability of the solution, or does not use the application provided, it is quite possible the duration of the contract will be curtailed. This is because the business partner simply will not renew the contract with a vendor that does not understand the requirements.

* What happens if the vendor solution fails?

Many cloud providers will mention their solution is based on a few data centres or locations; however, there are challenges when looking at the cloud-based system in South Africa.

In the South African market, the data needs to be stored in a local location, which can be based in a data centre. However, if the data centre goes down, the client will probably want to access the information from another data centre, which means the historical information needs to be replicated in another location.

Testing the worst-case scenarios is the most effective way to go about things.

Some providers also create direct connectivity links to data centres. If the data centre fails, those links will not keep the solution running, and it could also mean the MPLS or other solutions need to be rerouted to allow the client Internet access. Additionally, the solution provided needs to have open Internet-facing access for clients that need to make use of the business solution while their data centre is being repaired.

* Business continuity requires the solution be resilient and best in class

When looking at reliable cloud solutions, it must be understood what the impact is on the rest of the business. While many companies believe cloud systems will resolve their business continuity issues, this is not always the case. A mature business continuity plan looks at the whole business process, from location to disaster recovery and other measures, to ensure the business is able to function without all the optimal resources when faced with a crisis.

* Change control and notifications of maintenance

Clients will understand that to deliver a quality solution, there will be occasions when the vendor will need to upgrade software and do routine work on its server infrastructure.

Although the disruption to service might be minimal, it's always best to let clients know that some work is being done in the background. Most cloud providers, virtual servers, and other technologies, like software networking, are making these changes much simpler, and easier to plan for - the key is for the vendor to communicate with its client base.

The vendor must make sure it always understands the implications of its service to its clients. It might build a solution to isolate certain parts, to enable the vendor to provide only must-have services if there is an outage. Special functionality or applications need to be switched off to ensure the core solution is still available to all the vendor's clients.

Make sure to always prioritise paying customers. A vendor's trial customers or free customers need to be able to receive a reduced service to give those paying customers preference; this also includes capacity planning and making sure that if the vendor has a significant number of requests, its service is not affected.

The design and application of cloud services makes the solution attractive to companies with large requirements for business continuity and disaster recovery.

When the cloud providers follow best practice methods, and make sure the solutions they provide meet the clients' needs and requirements all the time, every time, these cloud providers will become the preferred solution.

Cloud solutions should just work, all the time, every time.

Share