The news is here. Click here to find out more about 3PAR's Dynamic Optimization - the tool our customer, TrafficBroker, is using to maintain their array's performance profile as they expand capacity.
« dw'Ville: A coder's coma | Main | dw'Ville breaking news: Larry Have no fear, Pillar storage is here! »
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00e553e34fa488330120a5761813970b
Listed below are links to weblogs that reference 3PAR Countdown: TrafficBroker maintains performance levels with Dynamic Optimization:
You can follow this conversation by subscribing to the comment feed for this post.
This is only a preview. Your comment has not yet been posted.
As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.
Having trouble reading this image? View an alternate.
I had to do a lot of dynamic optimization earlier this year.
When I created the file system originally I provisioned more space than we needed with a little assurance from our Exanet support rep that their file system was thin friendly.
Turns out at that point it was not thin friendly and it began expanding fast.. So I was forced to go from RAID 5 3+1 to RAID 5 5+1 and that saved about 12TB of space. Fortunately I started early enough that we didn't max out the system(a couple of times we exceeded 90% capacity). Right now we're running at about 112TB raw usage and about 58TB written space(in the NAS file system).
In the end it was roughly 100TB of data that I had to restripe. It did take more than 4 months to do it(our disks are very busy, and 100% SATA..), but there was no impact.
The most recent release of Exanet's code is thin friendly and they say can tie into the thin built in stuff, we plan to upgrade to it later this year.
Also expecting to add another 100TB of space early next year, and probably spend another 4-5 months re-striping that as well. Fortunately it runs by itself for the most part just have to check in on it a couple times a week.
Been hovering at around 85-87% of max for a while now, should be in the clear and able to last until we get the next round of spindles(more for I/O but extra space is welcome too, can store more snapshots and stuff).
If it was (almost) any other array I think we would of had to buy a bunch of disks to fix the situation and/or take a bunch of downtime.
Posted by: nate | September 16, 2009 at 03:26 PM
Marc - I hate to say it - but she does that way better than you do buddy!!
Posted by: Chris Fricke | September 21, 2009 at 02:49 PM
Gee, you think?
Posted by: marc farley | September 21, 2009 at 03:21 PM