Should We Move to Azure?

  • Comments posted to this topic are about the item Should We Move to Azure?

  • I think it would be an interesting case study. To be honest I'm surprised that you haven't already done it.

    As a private individual I'm sticking with the site as I have done as I've moved jobs throughout my career.

    As a corporate individual I'd need to make sure that emails still get through and that the site wouldn't get blocked.

    In terms of moving to the cloud I'm not worried about it per se I'm just in the mindset where a small system will fit on my existing tin without bothering core activity then why bother moving it? If it were a case of it consuming internal resources (CPU, RAM, IO) that could best be used for core business activity then I could see a case for it.

    Would you be talking about using SQL Azure or SQL Server in the Azure cloud?

  • I think that there are multiple ways to get true value from this venture and, by asking our opinion, you have started down this valuable track. I hope you continue to do so.

    This should be the source of great content for SSC. There can be articles, editorials and even polls all along the way. If you document the whole process then you will have a wonderful case study. As well as the technical implementations there are also the technical discussions to document. For example, I for one would love to read about how security was considered, in what contexts, what options were considered and why as well as the final solution. Publicly documenting security may feel like walking around naked but details may be redacted where necessary so I would be comforted to know that passwords were salted and asynchronously encrypted but I do not need to know the encryption algorithm.

    As a user I would accept disruptions to the service, particularly if we were warned about upcoming changes, as one of the points is that we are inside the process. This would have to be our accepted cost for all these benefits. I would also understand that certain site features may become temporarily duel hosted (i.e. different URLs) with one being read only during the transition process.

    Also, SSC can choose to engage this community at any level for any part of this project (if it happens). Do you discuss design or implementation decisions via editorials, for example, or even offer decisions up for review? This can only work if SSC is totally in control and is open and firm with the engagement at each point i.e. the responsibility lies with SSC so it must be clear that this is not an Open Source project but an open project.

    There is even some slightly off topic articles regarding web page generation to be garnered here. As well as a lot of kudos!!!

    I really hope that you guys do this.

    Gaz

    -- Stop your grinnin' and drop your linen...they're everywhere!!!

  • Trial and error....and success. How else do we evolve without trying new things? Cloud storage and computing is an excellent idea that is completely debatable. I say weight out your pros and cons. What's the worse that could happen? I can't view my subscribed threads and search for a solution to my head pounding problem.

    I say go for it.

  • DO IT!

    You have been jabbering about "Cloud This" and "Cloud That" for years. Since you really don't have any medical, financial or legally sensitive information and SSC is a web site, not a critical business service set in a "real" enterprise, you should be a perfect candidate for Azure. Practice what you preach/pimp.

    As for email, you need to tweak it anyway. I currently get two emails to the same email address, everyday. Figure this out, you might be able to reduce your emails. Also consider forcing people to renew/validate every couple of years.

    As for confidence, that's not an cloud issue. The primary cloud issues are security, control and connectivity. Many of us and our clients can't legally use it, don't trust it in light of the NSA revelations, want more admin control and don't fully trust the US network providers not to screw us over...

  • Please do so. It will be great to find out the realities of moving SSC to the cloud and whether we ought to do it ourselves.

    I am somewhat surprised that SSC is not in the cloud. It seems like a perfect cloud application.

    Tom

  • I also want to add that if you decide to go ahead with side-by-side versions then I am happy to be a guinea pig as I am sure that not everyone will both be able to and want to.

    Gaz

    -- Stop your grinnin' and drop your linen...they're everywhere!!!

  • SSC, in my humble opinion, should do it. As others have said, hearing about the process will be a tremendous learning experience for all. Plus, you'll be moving into a special place that few of us dare to go...on our own. (I'm not talking about corporate endeavors....) 🙂

  • Sure give it a try. Let us know the details.How to documentation will be good.

  • Please do and document the process. It will be very interesting.

  • SQL Server Azure offers scalability when it comes to storage, memory, and CPU, but I'm guessing that this website really isn't that massive and it grows over time in a predictable way. I'm guessing that your biggest concern should be how moving your websute and databases to the cloud would impact accessibility and latency. Latency; that's probably the game changer.

    "Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho

  • I would be interested in seeing the real world comparison of the differences in (a) operating costs, and (b) ease of maintenance, between the in house server vs cloud.

    But while I see it as an interesting exercise, I don't (yet) see a compelling argument for either hosting it on the cloud or keeping it on your own server.

  • "Or is SQLServerCentral just a service and you don't care how it works, as long as it does."

    For me, it's this. As we tell our own users, I don't care how it works behind-the-scenes, I care about the front-end and the content. That said, all my work is in a classified environment so "the cloud" has never, and will never, be a player for me. It's be an interesting read if you do go this route, but personally I could never benefit from lessons learned. Best of luck if you choose to do it!

  • I'd like to also see your planning for cloud failure. This would be very interesting for me. I have read about cases where a cloud provider goes out of business and end-users are given only a few days to pull everything off. Also an exploration of the legal aspects would be interesting. Right now my understanding of the law says that if the Feds grabbed the servers for an unrelated issue they wouldn't have to give SSC back their data. How will you prepare for that (and should you?). All very interesting. Another question might be now that you don't control your servers and data 100% will you do anything different. An admin at a cloud provider was recently fired and used some back doors he set up to cause trouble. Will you prepare for that? Will you save money? What's your ROI? Will you prep an ROI analysis before moving?

    I think you should do it (all of it).

  • No better way to learn the benefits or problems than try to try to implement a solution / application at production.

Viewing 15 posts - 1 through 15 (of 71 total)

You must be logged in to reply to this topic. Login to reply