Diomede Storage Blog

All Things Diomede.

DioCLI Replication Policy demo

Posted by diomedestorage on August 7, 2009

We’ve had several inquiries on how to use the Diomede Storage Replication Policies lately, so here’s a quick example and we’ll update the technical documentation shortly.

This example uses DioCLI (a command line interface to Diomede) to:

  1. Create a 2nd Replication Policy (the 1st having been created by default when the account was created)
  2. Set that new Replication Policy to be the default (so all new uploads will have this Replication Policy assigned)
  3. Upload a new file
  4. get the file properties for that file to see that the new default Replicaiton Policy was applied.

The new policy created specifies only that 3 copies are made on Diomede Offline Storage. So, what happens is all new uploads arrive at Online Storage, but then the Replication Policy is applied within minutes which:

(a) creates 3 new copies of the file on Offline Storage (for lower cost and power consumption)
(b) deletes the original 1 Online copy.

DioCLI example

Using DioCLI to create a new default Replication Policy

Posted in cloud storage | Leave a Comment »

Western Digital Ship 2.5″ 1 TB hard drives (PCWorld)

Posted by diomedestorage on July 27, 2009

Sweet! Small form factor 1 TB drives are on their way, and to be tested in new Diomede Storage servers shortly.

http://www.pcworld.com/businesscenter/article/169095/western_digital_ships_1tb_laptop_hard_drive.html

(Note: we never make any significant changes to production hardware without a comprehensive testing stage, so if these drives pass, it will still be several months until production data (your files!) are stored on them.)

Posted in storage industry news | Tagged: , | Leave a Comment »

Breaking news: storage is faster and cheaper

Posted by diomedestorage on July 21, 2009

Despite the sarcastic headline, we’re always overly excited to learn about new storage technologies:

Intel boosts speed, cuts prices of solid-state drives
http://news.cnet.com/8301-13924_3-10291582-64.html

Posted in commentary, storage industry news | Tagged: , | Leave a Comment »

Greening the Internet: How much CO2 does this article produce?

Posted by diomedestorage on July 13, 2009

Nice to see some mainstream media coverage of the major data center power consumption and pollution problems we are all facing:

Greening the Internet: How much CO2 does this article produce?

http://edition.cnn.com/2009/TECH/science/07/10/green.internet.CO2/index.html

Note: the text of that article (7,684 bytes) consumes only 0.01003 microwatts of power when backed up on Diomede (about 60 times less than a convention storage system).

Every little bit does add up (pun intended :)).

Posted in Uncategorized | Leave a Comment »

FreeDOS turns 15 today

Posted by diomedestorage on June 28, 2009

Neat – FreeDOS turns 15 today:

http://sourceforge.net/forum/forum.php?thread_id=3315853&forum_id=971311

I picked it up from here:

http://www.reddit.com/r/programming/comments/8wed3/freedos_turns_15_years_old_today/

I suppose we should release our DOS Diomede client, huh? :) (really)

Posted in commentary | Leave a Comment »

1996 Press Release

Posted by diomedestorage on June 23, 2009

I realize there are countless articles like this out there, but this one seemed pretty funny since it was “only” 13 years ago…

http://berkeley.edu/news/media/releases/96legacy/releases.96/14362.html

SIX terabytes! :)

Posted in commentary | Leave a Comment »

Diomede Storage and Cloudfront now both support request logging

Posted by diomedestorage on May 7, 2009

Amazon recently announced they have added request logging to Cloudfront.

http://aws.typepad.com/aws/2009/05/amazon-cloudfront-request-logging.html

This functionality, and more, is also available with Diomede Storage. Some of the related Diomede service calls to check out are:

  • SearchDownloadLog
  • SearchUploadLog
  • SearchLoginLog
  • SearchFilesTotalLog

Here’s a quick screenshot of these calls used from diocli (from a small test account I just created, all default parameters used for each call):

Photo and file sharing by Super Quick Link (www.superquicklink.com)

These are also the calls used to generate the graphs in DioMonitor:

Photo and file sharing by Super Quick Link (www.superquicklink.com)

Happy graphing!

Posted in Uncategorized | Leave a Comment »

Blow up a safe?

Posted by diomedestorage on May 4, 2009

Crazy, I didn’t think stuff like this really happened…

http://www.australianit.news.com.au/story/0,,25299140-5013040,00.html?from=public_rss

Posted in Uncategorized | Tagged: | Leave a Comment »

Archivd post about cloud vendor lock in

Posted by diomedestorage on April 10, 2009

Here is a good blog post on Archivd about cloud vendor lock in:

http://blog.archivd.com/1/post/2009/04/the-clouds-hidden-lock-in-latency.html

This is a good issue to raise and is an important reminder why you should choose a cloud storage vendor that provides easy mechanisms to physically export your data in bulk, avoiding any network limitations. Diomede’s bulk import and export functionality enables just this.

Diomede Storage is going to keep your business because we have the best service and the best solution for you – we’re not going to “hold your data hostage” for technical reasons or otherwise. If you ever need to get all your data out for whatever reason, it’s yours and we can ship it to you in bulk very quickly. Regardless of how much you love Diomede, we understand that things come up – acquisitions, partnerships, legal regulations, etc – that you’ll need to be able to handle.

As a side technical note: the author is saying this is due to network “latency”. While the latency to access “hosted clouds” (like Diomede) is higher than private clouds or services on your LAN, network latency is not specifically the reason it’s difficult to transmit large amounts of data to or from clouds – it’s typically due to the max throughput of the client (i.e. the size of their pipe to the Internet). It is possible to achieve high data transfer rates over high latency connections and a number of companies (one example) have developed protocols to tackle this problem. But regardless of the technical reason, it’s a good issue to bring up.

Posted in bulk data import, cloud storage | Leave a Comment »