Are we there yet?

We have been receiving heaps of mail regarding the Cloud Xtender beta release date, the continuation of Drive Bender v2 development and PoolHD v2. So I thought it best to touch base and give everyone the run down.

Cloud Xtender
The Cloud Xtender beta has been our core focus over the past few months, the interest in this product has been overwhelming… so we have been hammering away in an effort to get it out the door. We have a working version running in-house, and are now just fine tuning how Cloud Xtender interacts with different cloud providers. At this stage we are looking at mid November as the first beta release… and providing there are no show stoppers, the final release will happen shortly thereafter. Although we have cloud storage providers for AWS (Amazon) S3, Azure, DropBox, Google Drive and OneDrive (formerly SkyDrive) we are planning on having only AWS S3 and Google Drive in the first beta. Reason being is that we really want to get the first beta out the door… as it stands we are well behind our initial release schedule (hmm who would have thought!), so in an effort to get the first beta out, we have decided to exclude the other providers for the first beta. To give taste of what is to come, I have posted a demo video of an earlier build, this shows the interface, and how we create and add cloud drives etc. You can check it out here.

Drive Bender v2
This leads me into Drive Bender v2’s cloud duplication support. We are planning to continue with the v2 development when Cloud Xtender is sorted, mainly because the cloud duplication feature uses much of the Cloud Xtender code. I will also mention VSS, we are getting a small number of users asking what happened to this feature in v2. As a refresher, here’s what I wrote about this in a blog post last year…

VSS Support
I thought I would touch on this much requested feature first up. For those not familiar with VSS (aka shadow coping), let me point you to http://en.wikipedia.org/wiki/Shadow_Copy. With v2 we have implemented VSS support and have this under internal testing. We are able to snapshot a pool, and access these individual snapshots independent of the pool itself. This all works great, however one of the key requirements is external application support for doing this, known as requestor support (the requestor is the calling application such as Windows Backup). Again we have this working, however during testing we have run up against a number of issues. The most notable is reliability, for some reason the VSS requestor (i.e. Windows Backup), cannot always successfully snapshot the pool, and the entire backup fails with an error. Also the success of this seems to be very machine dependent, and we have no idea why… unfortunately good technical documentation on VSS very limited to non-existent. I should also mention that the internet is filled with stories of VSS failures… this is one finicky beast!

So what does this mean for VSS support? Well our original timeline had full VSS support in beta 1, and this was meant to be available mid August, however given the issues we have faced, this has not happened (obviously). At this stage we have decided to move on with the other stages of the v2 development and beta releases, while we continue to work on these VSS issues in the background. At the very least v2 will support pool snapshots, snapshot mounting and snapshot management. If requestor support continues to allude us, then we have a plan B (I will post on plan B when we come to a final decision).

We have revisited VSS a number of times since, only to run into the same lack of reliability… however we have not given up, and will keep plugging away when time permits.

PoolHD v2
Version 2 of PoolHD is also set to be released very soon. PoolHD v2, is based heavily on the Drive Bender v2 code (as v1 was), which is why we have been delaying… we had been chasing some odd permission issues in Drive Bender versions prior to v2.1.9.0, so we didn’t want to move forward with PoolHD v2 until these were sorted. Stay tuned!

12 thoughts on “Are we there yet?”

  1. Hi,
    are there any plans to create a provider for OneDrive for Business? Since Microsoft announced 1TB storage per user it would be greate if Coud Xtender could make use of this.

    Thanks

    1. Hi,

      We have plans to expand the providers as needed, as for specific providers such as OneDrive for Business, if there is a demand, then yes.

  2. Looking good Anthony!

    If you connect multiple cloud services, won’t you eventually run out of drive letters?

    1. Hi Keith,

      You can connect to multiple storage services using the same drive letter. While you can use Cloud Xtender to create a drive letter that points to a single cloud storage service (and given the performance, I suspect many users will use it in this way), one of it’s key features is to be able to pool multiple cloud storage services under the one drive letter… think of Cloud Xtender as cloud storage pooling.

  3. When you say “given the performance” what do you mean?

    I have tried three different competing solutions (the usual allowing to mount a cloud provider as a lettered drive) and for one reason or the other they weren’t suited to my needs.

    Two were god-awful slow, one had the bad habit of downloading the *whole* file when just browsing it triggered an automatic thumbnail generation from XBMC (which obviously doesn’t happen if proper file access is mimicked).

    What sort of speeds are we talking about? Say Google Drive, it fully saturates my bandwidth when accessing it from the webapp. What could I expect from Cloud Xtender? 🙂

    1. Cloud Xtender, when operating in backup mode (there are a number of different modes that effect how a file is stored locally), does not enumerate the cloud service when a users is looking at the local drive, it enumerates an image of the remote file system. This of course means speed is basically the same as a native drive. In addition, Cloud Xtender can cache small file block requests. For example, if you list a media file, Explorer may request certain meta data from that file, in such cases, only the requested file info is pulled from the cloud, and is then cached for then on.

      1. That’s nice to hear. Apart from browsing speed, I was interested in understanding download and upload speeds. As mentioned, two of your competitors (in the “cloud drives mounting” arena) were really, really slow. Like one twentieth of the available bandwidth in download.

    1. Not as yet… it is very much ready, however we have an issue the code for one of the cloud services. Once this is sorted and tested we’ll release it.

  4. Anthony, concerning VSS, why not provide a Drive Bender backup tool (possibly block level)? Wouldn’t that be easier? I’d be surprised that there are many other uses cases (except backup) for VSS with Drive Bender.

Comments are closed.