> On Aug 14, 2023, at 1:48 PM, Milo Velimirović <milovelimiro...@gmail.com>
> wrote:
>
>
>
>> On Aug 14, 2023, at 11:52 AM, Paul Koning via cctalk <cctalk@classiccmp.org>
>> wrote:
>>
>>
>>
>>> On Aug 14, 2023, at 12:42 PM, Chuck Guzis <ccl...@sydex.com> wrote:
>>>
>>> On 8/14/23 06:07, Paul Koning wrote:
>>>
>>>>> https://ia800809.us.archive.org/30/items/TNM_Control_Data_807-808_Disk_Files_-_Control_Dat_20170826_0047/TNM_Control_Data_807-808_Disk_Files_-_Control_Dat_20170826_0047.pdf
>>>>
>>>> That drive was the fastest transferring disk, by a large margin, for quite
>>>> a while.
>>>>
>>>> Didn't Cray use multiple bits in parallel disk drives in the Cray-1?
>>>>
>>> I recall that the first time I heard the term "RAID", it was from one of
>>> my friends from Cray. I thought it was pretty innovative.
>>
>> It's been a while, but I think that RAID originated in academia, though Cray
>> may well have been one of the earlier commercial users.
>
> This is probably one of the earlier papers on RAID that came from Academia
> from the group that invented it.
>
>
> https://www2.eecs.berkeley.edu/Pubs/TechRpts/1987/5853.htmlA Case for
> Redundant Arrays of Inexpensive Disks (RAID) | EECS at UC Berkeley
> eecs.berkeley.edu
> <touch-icon.png>
>
>
> "Increasing performance of CPUs and memories will be squandered if not
> matched by a similar performance increase in I/O. While the capacity of
> Single Large Expensive Disk (SLED) has grown rapidly, the performance
> improvement of SLED has been modest. Redundant Arrays of Inexpensive Disks
> (RAID), based on the magnetic disk technology developed for personal
> computers, offers an attractive alternative to SLED, promising improvements
> of an order of magnitude in performance, reliability, power consumption, and
> scalability.
>
> This paper introduces five levels of RAIDs, giving their relative
> cost/performance, and compares RAIDs to an IBM 3380 and a Fugitsu[sic] Super
> Eagle.
That fits what I was thinking about.
>> What I was referring to wasn't RAID but rather single drives with parallel
>> (multiple tracks concurrently) transfer, just like the 807/808. Does that
>> ring any bells?
>
> This sounds like it’s similar to, but not the same as fixed-head disk, or
> head-per-track disks. DEC had several models of relatively low capacity in
> the 1970s positioned as swapping devices with capacities from 128kB to 2048kB.
But all those, drums as well as fixed head disks, are serial devices. Not that
they need to be; given head per track it's clearly straightforward to do
parallel recording, but certainly the DEC devices do not do so.
> Or were you thinking of something more recent, like this:
> https://techmonitor.ai/technology/designs_disk_arrays_with_ncr_parallel_disks_with_cray_researchDESIGNS
> DISK ARRAYS WITH NCR, PARALLEL DISKS WITH CRAY RESEARCH
> techmonitor.ai
>
That seems to be the one, at least at the abstract level it matches what (very
little) I remembered. Is this a Cray 1 era article?
paul
