Recommended disk space

[andrewsears]

I'm not sure where I should ask this question, but we are looking to implement an instance of OJS, and I'm trying to get an idea of the amount of disk space that would be required to run it.

One journal, one issue, about 6-8 articles, and assuming there would be an average amount of reviewing and savings (versions and all).

Any help would be greatly appreciated, and thanks in advance.

Andy Sears

[jmacgreg]

Hi Andy,

I can't give any strict guidelines on this, as the amount of space required varies so greatly between journals. Generally speaking, our observance has been that space is cheap; and that it's far trickier to find, for example, a shared-host service provider that provides a decent PHP and MySQL environment (while these same service providers seem to take great pride in providing gigs of space).

You might be able to come up with a metric if you consider an average article at x MB (HTML and/or PDF); and also versions of that article in your preferred format (Word; rtf; whatever), multiplied by the number of versions you expect to be stored within the system (OJS versions a file as soon as it is submitted, so typically, you'll have 1 original submission file; x number of original supplementary files; and then x number of files for each review round; plus x numbers for each round of copyediding, proofreading and layout editing you'll optionally go through). Multiply this by the number of articles you'll expect to be submitted and vetted in a year, and you might have a reasonable estimate. But again, this number will likely be prone to some wide, wide variance.

Cheers,
James

[andrewsears]

James,

Thank you for your reply.

The server setup and configurations won't be a problem, and really neither would size. I am currently writing a spec for our university, and we just need to fill in some blanks (e.g., expected size after 5 years). So I was hoping that someone might have a rough number to expect for one issue. There will definitely be tons of factors that go in each issue, but would it be safe to say that 20 MB for a 6 article issue (with a handful of submissions and reviews per article)? 50 or 250 MB?

Or have people seen that with lots of Word document uploads, it can get to almost 1 GB per issue?

Regards and thanks, Andy Sears

[mj]

Hi Andy,

I understand your need to have an actual number, but I will most definitely stress James' comments above -- these can vary widely from issue to issue, year to year, journal to journal. So, take anything here with a very large grain of salt.

That said, I can make a few conservative guesses based on my experience from a medical journal I used to work with. They published about 50 articles per year, submitted in MS-Word, with about 3 review versions, and then published in HTML and PDF. One of the things that are difficult to account for that can take up space quickly are large image files and supplementary files (eg. video, data sets), so those can confound things somewhat. Assuming Word files are around 1MB and the corresponding publication PDF is around 2MB (both on the high side, I'd reckon), you'd be looking at about 7MB per article, which for a 6 article issue would round up to 50MB. Off the top of my head this sounds entirely reasonable, ie. growing at about 200-500MB/year, but you may want to build in additional padding.

Hope this helps,
MJ

[andrewsears]

That does, thank you both James and MJ.

I totally get that the range could really be all over the place. And we know in practicality that if we need more space, we can always just add more. This is mostly just for paperwork stuff.

Again, thank you both.