Looking for advice on migrating to Cloud

[matt beamish]

Hi folks,
I work in an organisation where I have developed dozens of Access databases over the last 18 years or so most of which remain in operation and many are core to our daily operation. The time has come to really push to move the back ends from the standard network storage on which they have always sat, to something that is cloud based, is faster & will enable wider access including from mobile devices. I need advice on what options to go for and what the differences are between available systems (SQL Server, Azure, Sharepoint, .net front ends etc). If the front ends will need rewriting due to a back end change then this is a good time to contract out and so I would be looking for a UK based professional to help with this.
thanks
Matt

 

[theDBguy]

Hi Matt. Going to the cloud may not mean the app will become "faster" as you had mentioned you may be expecting to happen. If you want to continue using your existing Access front ends, you can look into storing your data in SQL Server or Azure (or even MySQL or Oracle or DB2). But if you're planning on using a browser or creating a mobile app to access the data, then you'll need to look for something else other than an Access front end.

 

[Mark_]

Something to remember about "Cloud" storage; your data still resides on a server (or servers), but they are often no longer under your direct control. This may require a rework of your local network topology to keep speeds up to acceptable levels. It also means that you may need to rework your applications to move as much data processing to the back end as possible.

Your first step should be verifying reliability and throughput between your location(s) and where your data would be stored. Finding out you have slow connection is something you need to identify first before deciding on a hosting solution.

 

[Gasman]

If this ever takes off, you might be able to use this?

https://azure.microsoft.com/en-gb/services/virtual-desktop/

 

[Uncle Gizmo]

Also search the forum for Citrix, there's been a lot of posts about it and you might find it a suitable option.

Sent from Newbury UK

 

[CJ_London]

you do need to be very clear what the reasons are for moving to the cloud. access from anywhere, access by partner businesses/customers/consumers are good potential reasons, speed is generally the bottom of the list.

If it is a speed issue, look at citrix/terminal server perhaps using sql server/express back end but retain your access front ends (on citrix/terminals server) with minimal changes (assuming it is already optimised for speed). This should perform pretty much like your backends are on your local drive. Should also solve the problem of connecting via wireless and from non office based locations and from multiple types of devices. They should also handle the additional heavy security burden from being on the cloud.

I know you can connect to terminal server from apple devices (including iphone), but don't know about other OS's

However your front ends will no doubt need to be redesigned to fit smaller screens.

SQL Azure works well when the devices connecting have a known IP (i.e. from the office), not so good if you want to connect from your local coffee shop. But other than taking advantage of the additional BE functionality, only small changes to the BE. Be aware you do need to pay for a relatively high level of service otherwise performance will be slow.

After that you are into brand new apps with your existing access apps being used purely to demonstrate 'it needs to work something like this'. To my knowledge there are no apps out there to quickly develop web based front ends aka Access beyond the most basic requirements. You will probably be into using something like Visual Studio or NetBeans.

Before you start I recommend you get a budget together of what moving to the cloud is worth to you. You may find there are apps already out there which can perhaps be tailored to your specific requirements, either through settings or some bespoking. Either way there will be a cost.