4

Certain journaled file-systems do suffer from performance penalties for hosting the journal on the same drive as the filesystem itself. Using an external journal can improve write speeds. With the advent of SSD's, it strikes me as entirely possible that a single SSD could support multiple external journals for filesystems hosted on traditional rotational magnetic media. While the Anandtech "SSD Anthology" did cover some of this, they didn't cover the "multiple journal" test-case.

Journals tend to be pretty small, so even a small, fast 32GB SSD could provide a lot of speed for multiple large file-systems. It would probably wear much faster than general I/O patterns out suggest, which is why having a larger device than you'll end up using is a good idea.

Have any of you done something like this, or even just used an external journal? I'm curious about real-world cases.

Improve this question
1
  • I really like the idea of this, but I'd like to see results from appropriate tests. Cool idea, though!
    – Matt Simmons
    May 30, 2009 at 3:01

3 Answers 3

Reset to default
2

Given that the amount of data written to the journal is usually very compact and most server systems (at least those where performance matters) come with RAID controllers with a Battery Backed Write Cache, the penalty of using a journal is mostly a non-issue. The possible speed improvement is very much offset by the added complexity and problems of having an external journal (think setup, maintenance, (disaster) recovery, increased error rate from less used code paths, etc.).

This is probably one of the main reasons people hardly venture down this road.

Now for one novel supporting use of SSDs, when it comes to filesystems, would be ZFS L2ARC, but that's a whole different story.

Improve this answer
1
  • My goodness, that's a pretty thing. And yes, another topic all together.
    – sysadmin1138 ♦
    May 30, 2009 at 1:45
2

Most of the current SSD technologies are much slower at writes than they are at reads so I'm not sure that using them for a journal would improve speed. The major storage vendors seem to be using this technology as an additional layer of read cache due to their fast read and low latency. This is what ZFS is doing with L2ARC and this works quite well with certain workloads.

But since storage workloads tend to vary a lot how much of a difference having a larger read cache has in any situation will depend on the access patterns and size of the working set of data.

Improve this answer
1
  • Speed is one thing, latency is another. SSDs may not be much faster than HDDs during streaming writes, but they will have lower latency.
    – Hubert Kario
    Sep 12, 2012 at 20:54
0

You also need to consider the risk that the SSD's have a shorter lifetime than hard drives. A good benchmark on hard drives is look at the warrantee. A good server hard drive will be 5 years, most SSD's are 1 year. The reason for the shorter lifespan is the SSD cells will wear out eventually (you only get so many write cycles). Although most drives have firmware that evenly divides where the data is stored to increase the lifespan by spreading the love across all the cells.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .