I currently have 2 VMAX 20Ks. I have a my SSD separated to specific pools.
I do have one mixed pool (SSD and Spinning). This is a response I got when
I posted this question:
If a Storage Group contains a mix of SSD and HDD volumes, to divert datasets
on HDD volumes, assign them a Storage Class with these attributes:
DIRECT MSR = 1
DIRECT BIAS = R
Do not specify any values for SEQUENTIAL MSR or
BIAS. *
To direct storage allocation away from SSD, specify a DataClass with:
DIRECT MSR = 10
DIRECT BIAS = R
In case of HSM restore, dataset will be allocated depending on its SMS
attributes
So, you can separate datasets in a mixed pool by creating classes in DATA
and STORAGE with FAST and SLOW attributes.
HTH
Lizette
> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]] On
> Behalf Of R.S.
> Sent: Thursday, March 06, 2014 7:33 AM
> To: [email protected]
> Subject: Re: SSD in DFSMS
>
> W dniu 2014-03-06 15:19, Lizette Koehler pisze:
> > What hardware are you using SSD in?
> >
> > IBM DS8???
> > EMC VMAX or DMX
> > HDS
> >
> > If you have a VMAX 40K then the product FAST VP probably will replace
> > the need for you to tune the SSD in HSM.
> >
> > If you have a DS8870 then Protect Tier or Fast Tier may replace the
> > need to manage the SSD.
> >
> > The Hardware Storage Arrays are now taking over moving data that is
> > "hot" to faster devices without HSM or the user knowing it is occurring.
> >
> It's VMAX (first model), however my quesion was rather general, just to
learn.
> I'm aware of FAST VP or similar features, but we're not using it.
>
> --
> Radoslaw Skorupka
> Lodz, Poland
>
>
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
