🚀 go-pugleaf

                Vax Cluster Storage Solutions
                =============================

        I am writing to seek information about the new DEC STORME system
        and DEC RAID directions.  We have at Monash University many Vax
        systems 64xx, 65xx and are soon to take delivery of a 66xx system.

                As far as disk subsystems goes we have made use of the
        HSC40/50/70 technology with RA70/71/72/81/82/90/92, Emulex, MTI,
        and NEC SDI drives. We have a DEC MPP with a parallel disk array
        together with numerous Vaxstations and Decstations.

                With our large system workload increasing we are looking
        down the track to try and find what new storage solutions are likely
        to be available in the near future. The HSC90s with Cache may not
        be the only way to go. We are also looking at ESE50 solid state
        disks as a possible solution.

                I am looking out for equipment that will replace the HSCs
        or co-exist with them on the CI Bus, that is, connected to the Star
        coupler directly be faster in terms of I/O requests /sec and have
        some measure of redundancy maybe a RAID solution.

                The best information I have at present is that MTI has a
        system called Failsafe and another called the Stingray, also CMD
        technology is working on a RAID 5 system that will interface to
        the CI bus.
                Have these Companies made agreements with DEC to interface
        on to the CI bus, or would we be likely to see some legal issues
        comming to the surface as they have in the past ?

                Are there any other manufacturers out there that are working
        in this area ? Is DEC itself looking at RAID solutions ?  Just how
        does a parallel disk array appear to VMS ? Is it just one big logical
        drive or can it be made to appear as several smaller logical drives?

                What happens in the Backup area?  If chunks are distributed
        across multiple spindles is backup in the traditional sense meaningful?

                If more than one element in the array fails and the RAID
        system is not able to reconstruct the data what is the recovery time?

                The purpose of this posting is an information gathering
        exercise and should not be considered as the direction the University
        Computer Centre is planning to move towards. I would appreciate
        hearing from any site or anyone with experience with the products
        mentioned above or any answers to the questions posed.

                                Thanks in anticipation,
                                        Russell Keil
                                         Vax Engineering.

                        email:-     Russell.Keil@cc.monash.edu.au