This is my situation...

Cloud Hosting - - disk I/O for my wordpress website (aka a database-driven website) is abysmal. But what's good about it is, there's no single point of failure, so very minimal risk of downtime.

So I thought I would so, local storage instead of redundant SAN offered in the cloud in order to raise performance. But then gain, this brings in the single point of failure thing (host doesn't offer RAID for local storage in cloud) + risk of data loss.

As you all know, Dedicated physical server again has a single point of failure. So, what do you think I should do? Any ideas are welcome.

EDIT: I meant, what should I do for disk performance in Cloud Hosting, without risking data loss?

closed as not a real question by mailq, John Gardeniers, andol, Ward, Bart De Vos Oct 5 '11 at 10:55

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, visit the help center. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    This question doesn't make a lot of sense. – jdw Oct 1 '11 at 19:18
  • @sysadmin1138 has identified the information you need to provide in your question so people can provide good answers. Please edit your question to include your risk acceptance, whether "slowness" is considered downtime, and what period of downtime is acceptable. Also what specific type of "Cloud Hosting" are you using? "Cloud Hosting" is so general of a term as to almost be meaningless. – kce Oct 1 '11 at 22:07

When assessing acceptable risk, one of the first things you need to determine is how much is at stake. What do you lose when your Wordpress site is down (corollary, is slowness considering 'downtime')? The next question to answer is how long you can be down while you recover.

Once you have those two answers, it will better enable you to judge the risks of various actions.

  • Always-up but slow is acceptable if downtime must be avoided at all costs
  • Almost always up but fast is acceptable if some downtime can be tolerated

In the second case it is critical that you actually have a recovery mechanism of some kind, and test it once in a while. Database driven websites generally can be recovered pretty well, you just need to be sure your DB backups are consistent and your file backups updated whenever file-level changes are made.

  • any suggestions for good cloud hosts @sysadmin1138? – Aahan Oct 1 '11 at 20:39
  • @Aahan: Shopping questions are considered off-topic. They are far too localized. – kce Oct 1 '11 at 22:21

One option to improve disk performance in the cloud could be to buy a cloud hosting package with enough ram in it that you could then use the ram to create a ramdisk, setup an async raid (write mostly) with the cloud storage and ramdisk & enjoy rather fast read speeds, some data resiliency... and likely a rather large monthly bill.

A ramdisk of course loses the data you store in it when the cloud VM/instance/server is restarted, so the async raid will help you retain your data. Though after every reboot the raid array will have to rebuild and performance will be slow until the rebuild is done...

Not all SAN's offer horrible disk performance btw, so you likely just need to shop around a bit. You are also likely aware of the cache plugins that can improve wordpress performance, but I suppose it should be mentioned that you may want to look into something like w3_total_cache, etc.

If you are concerned about data loss, buy from a host with a decent backup plan. Some of them even offer hourly backups.

  • Any suggestions for good cloud hosts? – Aahan Oct 1 '11 at 20:39

Not the answer you're looking for? Browse other questions tagged or ask your own question.