[AccessD] To the Cloud or Not to the Cloud. that is the Question

Gustav Brock gustav at cactus.dk
Fri Mar 28 02:36:30 CDT 2014


Hi Bill

You don't have to "ask for a data pump". You just connect to your data. SkyDrive/OneDrive is accessible as any other network folder, and SQL Server (nearly) like an in-house server. Outlook on-line acts like an Exchange server which you can interact with like any decent mail server (POP3, SMTP, IMAP).

This means you can copy or backup your data as to your requirements beyond whatever backup Microsoft claims to run internally.

/gustav


-----Oprindelig meddelelse-----
Fra: accessd-bounces at databaseadvisors.com [mailto:accessd-bounces at databaseadvisors.com] På vegne af Bill Benson
Sendt: 27. marts 2014 23:18
Til: Access Developers discussion and problem solving
Emne: Re: [AccessD] To the Cloud or Not to the Cloud. that is the Question

I have no experience with the cloud and perhaps in less open minded circles this negates the value of my opinion. But never this group, for which I am grateful. I would weigh the possibility of a web service outage which you are unlikely to have any control over, against the very much more expensive redundant hardware bases you will have to cover to even touch the reliability of the web.

The option to have the data available and accessible where and when you need it seems to be a no - brainer in favor of a web platform from a reliable cloud operation. And unless you are running heaps of insertions with huge datasets, I have little doubt you will be that unhappy with the performance. This is, if you are satisfied with MS as a development platform.

I do have to ask, have you looked into portability / migration if the tool proves unsatisfactory in the long term? Can you ask MS for a data pump version of your database or would you just have to turn it off and walk away, subject to a few exports of tables, but leave all your development behind? (Not that things would be all that different necessarily with an Oracle or SQL Server application either but I do think with those you can export the schema, no?)

I am sure if you don't get your feet wet trying a web service database it will be a mark against you career wise so why not use the current need to be able to say been there done that on someone else's dime. It comes well supported and at least fairly well recommended, do you have that much to lose?

I wish the future was nit web development because I am so uninitiated in it, wish I had an employer asking me to test out the platform Mark, I would say go for it.
On Mar 27, 2014 5:54 PM, "Gustav Brock" <gustav at cactus.dk> wrote:

> Hi Brad
>
> Your observation is correct. The Azure setup and the speed at which it 
> evolves is impressive and proves to me the old saying "don't ever 
> underestimate Microsoft".
> Because Microsoft didn't invent this, the "cloud". Amazon did that. 
> But having seen the light, there's no doubt that massive resources 
> have been allocated this infrastructure and the support it needs 
> beyond establishing the large and reliable data centers: Easy 
> administration targeted admins at enterprises, scalability not 
> possible with in-house hardware, programming interfaces, and - perhaps the most important part - the integration to ...
> well, everything you can think of.
>
> One of our clients has asked for a customized document store and flow 
> control. Problem is that this is a TV production company were all 
> except accounting use Macs, and the network setup is handled by 
> various Mac "techies" (you guessed it, it's a mess) so setting up a 
> server in this environment is perhaps not then at least the last option.
>
> So, seriously, I consider storing all data at Azure. JC is right, that 
> internet connection reliabilty and speed is a valid concern, but 
> clients like these have fiber connections at 100+ Mb/s. It will work. 
> Deciding to do so will, first, take very little to create the 
> development environment and, second, deployment will be close to just 
> "flip a switch". Add to this the establishing cost of zero money and 
> ongoing costs that without in-house techies are impossible to match.
>
> Still, I believe, we will arrange for an off-line backup of all data - 
> just in case.
>
> /gustav
>
> ________________________________________
> Fra: accessd-bounces at databaseadvisors.com < 
> accessd-bounces at databaseadvisors.com> på vegne af Brad Marks < 
> BradM at blackforestltd.com>
> Sendt: 27. marts 2014 21:37
> Til: Access Developers discussion and problem solving
> Emne: [AccessD] To the Cloud or Not to the Cloud. that is the Question
>
> All,
>
> A few days ago, I posted a question regarding a new database that will 
> be needed for a new small application (perhaps using Access 2007, 
> MySQL, or Microsoft SQL Server).  I received a number of replies to my 
> question and I appreciate the insights that were shared.
>
> In the mean time, I have also been doing some R&D work with Microsoft 
> Azure SQL Database.
> So far, I have been impressed.
>
> Yesterday in less than one hour, I was able accomplish the following - 
> Set up a new Microsoft Azure Account Set up a new SQL Database Then 
> with a small Access 2007 application using ODBC and Pass-through 
> Queries I was able to Add a new table to the new test database Insert 
> rows into the new table Retrieve this data.
>
>
> My background is in mainframe databases (Primarily IBM's DB2 and
> Cincom's Supra).  I can remember how much work it was to accomplish
> these same simple steps with these databases.  Sometimes we would spend
> days, just wrestling with DB2's security.
>
> It is my understanding that other Access users are using Azure with some
> success.  It also sounds like the costs for our use of Azure will be
> very small ($5.00 - $10.00 per month).
>
>
> At this point in time, it seems like this is a good option for our
> little project. (We don't have Microsoft SQL Server installed
> "in-house").
>
> I am curious if others have looked closely at Azure and decided to not
> go down this path.  Perhaps there are downsides that I have not yet read
> about. Perhaps I am missing some important considerations.
>
> Thanks,
> Brad 



More information about the AccessD mailing list