[Dick Ott]

“They had a good reputation from the people I talk to”
I don’t think anybody accused them of having a bad rep; remember, they were a start-up, not-for-profit company, and good customer service – at any cost – was vital to maintain growth.

“they couldn’t be all that bad”
From the hardware POV there are only so many ways to store data; the AS controller was made by Accusys and they are nearly identical to the other dozen Taiwan-based companies – but they have really only had that level of RAID controller for a few years as compared to Promise and Infortrend.
You have to consider experience: other companies have years of engineering growth and millions of units shipped, AS could not have had more than a few thousand unit shipped.

“AS went to Rorke data”
The Rorke Galaxy line was Infortrend a few years ago and if you look at the box closely (I did at NAB) it is now Promise.

1. AS was slower, proven by many users, and their hw failure rate was higher (are you talking HDD? controllers? or what?)
No storage vendor can be responsible for HDD failures other than using a model not suited for the duty (desktop drives in a 24/7 subsystem) or not handling the myriad ways in which drives can hose the controller.
From what I saw and heard from a few re-sellers there were several controller failures that brought down the surviving ‘good’ controller, and even some cases of data loss.
Also stability issues in larger configs where adding or removing a client caused the Xsan to freeze as the LUNs became inaccessible.
I have personally witnessed the AS-to-dumpster process, not kidding.

2. AS only worked with OS X
And not perfectly; they appeared to have a problem with SAN state changes where new clients could not see the LUNs until the AS units were rebooted. This was not a switch or client config issue, purely AS controller firmware.

3. AS was cheaper than Promise
Not significantly, and many deals were closed by giving huge margins to undercut the competitors to help generate business and ‘buzz’; remember they were a start-up and not-for-profit.
The price was competitive up front but would drop to close the deal – no profit there when you sell below cost.

4. What were some of the significant unresolved bugs?
See 1 and 2.

It is academic now as they are NLA and there will not be any support available at all.
Bear in mind that all of the shortcomings could have been addressed given time, but that wont happen now…

[Dick Ott]

Sorry but their storage was *not* faster than Promise’s – proven by many users that actually followed the published best practices – and their hardware failure rate was far greater than that of Promise and others. And Promise VTrak will work with any OS, not just OS X, right out of the box without tweaks and groans.
I have several friends that ended up with AS due to the cost only and by late last year had the chassis out to scrap and the HDDs re-purposed; AS could not resolve their SAN issues and always responded with “it appears to be something you did on the SAN”.
And they were going to retire the 16-bay unit – when it still has significant bugs that have not been resolved.
To be fair, they were a start-up and everybody has growing pains at that stage and it is too bad they did not have the opportunity to mature…

[Dick Ott]

Brendan, there is not really much tech to pick up.
AS was a “bundler”, the RAID controllers/firmware was done by a company in Taiwan – Accusys – and the ‘server’ was Lunux + Stornext.
The only real IP they had was the new mPath HW and SW.

The same went for Apple – no ‘storage wizardry’ there – the controllers were made by Areca, a Taiwanese company, and they were not ‘enterprise’ by any stretch with no write cache mirroring.

You may have been better off with a company that owned all their own IP and had been around for more than three years, like Promise…

[Dick Ott]

Yup, sales and support people at AS getting offers from Q.
They were using Stornext and Q can pick up the customer base.
So maybe not fatal news for Bob – unless he doesn’t like Quantum…

[Dick Ott]

It appears to be true Bob.
I heard from people that work there (small world, we all know each other) last week that they were looking for jobs; draw your conclusions.

I hate to see anybody lose their job that way (I have been in a few start-ups that went thud) and it is a bad position for the customers that embraced the company and products.

If by “Rolls Royce” you mean “slow and solid” then you may be correct; but their stuff was hardly the fastest or most robust storage for M&E.
At the end of the day it is all 1s and 0s and silicon and only so many ways to do it. You like the company, people, products, experience is varying degrees when you make choices like buying tech…