I have heard of Panasas on occasion and for some reason recently I saw a story or a link to them so I decided to poke around to see what they do. I like technology..
Anyways I was shocked to see their system design. I mean I’ve seen systems like Isilon and Xiotech and Pillar who have embedded controllers in each of their storage shelves, this is an interesting concept for boosting performance though given the added complexity and stuff to each shelf I imagine can boost the costs by quite a bit too I don’t know.
But Panasas has taken it to an even further extreme, putting a disk controller for every two disks in the system! I mean I’m sure it’s great for maximum performance but wow, it just seems like such a massive overkill(which can be good for certain apps I’m sure). I was/am still shocked 🙂
So today I was poking around again at the latest SPEC SFS results for NFS, and saw they posted some numbers finally.
Fairly impressive numbers but I just can’t get past the number of CPUs they are using. They posted 77,137 IOPS with 160 disks hosting NAS data (80 SATA and 80 SSD). They used a total of 110 Intel CPUs (80 1.5Ghz Celerons and 30 1.8Ghz Pentium Ms) and 440 gigabytes of RAM cache.
By contrast, Avere which I posted about recently (never used their stuff, never talked to them before), posted 131,591 IOPS with 72 disks hosting NAS data(48 15k SAS, 24 SATA), 14 Intel CPUs(2.5Ghz quad core, so 56 cores) and 423 gigabytes of RAM cache. This is on a 6-node cluster. This Avere configuration is not using SSD (they released an SSD version since these results were posted)
The bar certainly is being raised by these players implementing massive caches. NetApp showed off some pretty impressive numbers as well with their PAM last year, more than 500GB of cache(PAM is a read cache only) though again not nearly as effective as Avere since they came in at 60,507 IOPS with 56 15k RPM disks.
Hi Nate. Thanks for checking us out. A few comments on your observations:
Unlike Pillar, Xiotech, NetApp, etc., the Panasas ActiveStor (PAS) is a single namespace, parallel object store, so you’re comparing apples to bowling balls. Our customers use PAS in highly complex computational environments like aerospace, energy exploration, bioscience and in research institutes. Our customers use Panasas when throughput and scalability are paramount. As an example, we are the storage behind the world’s fastest supercomputer (> 1 petaflop) at Los Alamos National Labs, currently supporting over 1 PB and growing.
Our architecture is optimized to support massive torrents of scientific/computational data through our implementation of the emerging pNFS (Parallel NFS) standard, which will be part of in NFS v4.1. We also support NFS, CIFS, NDMP, etc. for more mundane workloads like home directories, backup & recovery and archival applications where 77,000+ IOPs is more than enough to support those workloads.
Finally, and I’m sure you’ll find this amazing, despite all the processors Panasas is very price competitive compared to the others mentioned on a dollar per GB basis and, Panasas is a great deal on a dollar per IOP basis, so don’t be intimidated by all of the horse power. It’s not overkill.
Comments are welcomeed.
Comment by Tom — March 12, 2010 @ 2:04 pm
thanks for checking us out! Yeah I know they are different, mainly Xiotech/Pillar etc are block only not file based systems. Your architecture is similar to Isilon though right?
And yes 77,000+ IOPS is more than enough for most workloads I’m sure!
Comment by Nate — March 12, 2010 @ 2:58 pm
some bowling balls are heavy and i accidentally dropped one on my foot. it is quite painfull”`.
Comment by Owen Bell — August 11, 2010 @ 9:05 pm
bowling balls are dangerous on the foot if you mishandle it.-;`
Comment by Joshua Taylor — September 30, 2010 @ 12:48 pm
bowling balls are quite dangerous to the hands of a newbie and untrained bowling player”-,
Comment by Oscilloscope — October 18, 2010 @ 1:02 am
bowling balls that are coated with an acrylic clearcoat are the nice ones”
Comment by Carpet Tiles · — November 7, 2010 @ 2:01 pm