[Info-vax] Databases versus RMS

[Dirk Munk]

Johnny Billquist wrote:
> On 2012-04-19 14.53, abrsvc wrote:
>> On Thursday, April 19, 2012 8:37:52 AM UTC-4, Paul Sture wrote:
>>> On Wed, 18 Apr 2012 18:08:52 +0200, Dirk Munk wrote:
>>>
>>>> Many hours, sometimes even a few days. Usually the disks are not
>>>> powered
>>>> by the batteries, only the RAM memory of the cache.
>>>
>>> I was given a demo of a RAM disk with real disk backup in the early
>>> 1990s. The battery in that was supposed to last long enough to copy the
>>> RAM contents to the physical disk in the event of a power fail.
>>>
>>> I never saw any of these deployed anywhere though. I wouldn't be
>>> surprised if they were too expensive.
>>>
>>> --
>>> Paul Sture
>>
>> Digital did work on SSD devices (Solid State Disk). I recall working
>> on the performance testing of these. The prototype for testing
>> simulated an RA81 disk and actually was breadboarded using the front
>> panel of the disk. I recall laughing at the drive's requirement to
>> "spin up" before it was available.
>
> DEC did SSD long before that.
> Anyone remember the ML11? It's a SSD for the Massbus, looking like any
> other massbus disk (well, apart from the fact that every massbus device
> have it's own identification, so you can see what it is, and thereby
> know the capacity of the device).
>
> Johnny

That's right. We had a similar device connected to PDP11 at the 
newspaper I worked at that time. But these were not SSD's, they could 
not store data without power present. They were RAM disks.

And indeed there also was an SSB device with a RAM disk and a real disk 
as backup. It had a battery to be able to write the data to disk in case 
of a power failure.