Clustering - how can I....... - VMS

This is a discussion on Clustering - how can I....... - VMS ; Phillip Helbig---remove CLOTHES to reply wrote: > In article , Gremlin > writes: > >> Phillip Helbig---remove CLOTHES to reply wrote: >>> In article , Gremlin >>> writes: >>> >>>> I have 3 Vaxen clustered (alright, 2 alphas and a ...

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 21 to 27 of 27

Thread: Clustering - how can I.......

  1. Re: Clustering - how can I.......

    Phillip Helbig---remove CLOTHES to reply wrote:
    > In article , Gremlin
    > writes:
    >
    >> Phillip Helbig---remove CLOTHES to reply wrote:
    >>> In article , Gremlin
    >>> writes:
    >>>
    >>>> I have 3 Vaxen clustered (alright, 2 alphas and a PersonalAlpha, but I
    >>>> am still old fashioned enough to describe them as Vaxen). Their purpose
    >>>> in life is to run WASD. I want the higher-availability of a cluster,
    >>>> the ease of management of single UAF etc.....
    >>>>
    >>>> So, the data source for the web server is an NFS file location stored on
    >>>> a Windows (sigh) server using RAID5.
    >>> WHY?!?!?!?!?!

    >>
    >>
    >> Because the cluster is of DS10Ls, using local IDE storage and the only
    >> simple, cheap way I can have a single source for the web server (I
    >> think) is to have it come from an NFS source. Why Windows? I have it
    >> running already and I am happy to not add another OS to the mix....

    >
    > You're talking about a cluster. Have the "single source" be a shadow
    > set consisting of disks on 2 or 3 cluster nodes. I admit I have no
    > experience with IDE storage---will HBVS work there?
    >

    The data source size is greater than the IDE disks inside the DS10Ls -
    and yes, this is on a hobbyist license. As the Windows servers is
    already running on a RAID set, the single data source with adequate
    redundancy is provided by NFS serving from Windows.

    I will experiment with the Failsafe-IP, it sounds like what I want - I
    am not after dynamic load balancing, just the ability for inbound reads
    from the webserver to hit a live Alpha without having to change the
    inbound IP address if a server crashes (unlikely with trusty VMS!)

  2. Re: Volume sets - yes, they can be very useful.

    Phillip Helbig---remove CLOTHES to reply wrote:
    > [...snip...]
    >
    > That should be "shadow set", not "volume set". A volume set is a
    > different beast. (Yes, volume sets can be combined with shadow sets.
    > In general, in these days of big disks, no-one needs volume sets
    > anymore.)


    Maybe in general, you're right. But at one of my customers,
    we're into volume sets in a big way.

    Like, 15 x 300 Gbyte disk volume sets.

    In another compute farm of 12 DS15's, each with a 2 x 300 Gbyte
    disk volume set. We can drive these farm members at 100% CPU,
    reading from one member, while simultaneously downloading a ca.
    200 Gbyte file (typical size) to the second member (yoyo-style).
    The only time these systems are ever idle is Monday mornings,
    since the daily "traffic" files for analysis are smaller at
    weekends (you can guess what business this is in). If these
    downloads are to the same disk as the one with the current
    analysis, we become (severely!) limited by disk-head bouncing
    back and forth.

    Volume sets are an absolutely perfect solution for this.

  3. Re: Volume sets - yes, they can be very useful.

    On May 18, 11:16 am, "R.A.Omond" wrote:
    > Phillip Helbig---remove CLOTHES to reply wrote:
    >
    > > [...snip...]

    >
    > > That should be "shadow set", not "volume set". A volume set is a
    > > different beast. (Yes, volume sets can be combined with shadow sets.
    > > In general, in these days of big disks, no-one needs volume sets
    > > anymore.)

    >
    > Maybe in general, you're right. But at one of my customers,
    > we're into volume sets in a big way.
    >
    > Like, 15 x 300 Gbyte disk volume sets.
    >
    > In another compute farm of 12 DS15's, each with a 2 x 300 Gbyte
    > disk volume set. We can drive these farm members at 100% CPU,
    > reading from one member, while simultaneously downloading a ca.
    > 200 Gbyte file (typical size) to the second member (yoyo-style).
    > The only time these systems are ever idle is Monday mornings,
    > since the daily "traffic" files for analysis are smaller at
    > weekends (you can guess what business this is in). If these
    > downloads are to the same disk as the one with the current
    > analysis, we become (severely!) limited by disk-head bouncing
    > back and forth.
    >
    > Volume sets are an absolutely perfect solution for this.


    would not be better off with striped disks? - there is a vms layered
    product that does this

  4. Re: Clustering - how can I.......

    In article , Gremlin
    writes:

    > The data source size is greater than the IDE disks inside the DS10Ls -
    > and yes, this is on a hobbyist license. As the Windows servers is
    > already running on a RAID set, the single data source with adequate
    > redundancy is provided by NFS serving from Windows.


    OK.

    > I will experiment with the Failsafe-IP, it sounds like what I want - I
    > am not after dynamic load balancing, just the ability for inbound reads
    > from the webserver to hit a live Alpha without having to change the
    > inbound IP address if a server crashes (unlikely with trusty VMS!)


    Yes, it sounds like just what you need.


  5. Re: Volume sets - yes, they can be very useful.

    IanMiller wrote:
    >
    >> [...snip...]
    >> Volume sets are an absolutely perfect solution for this.

    >
    > would not be better off with striped disks? - there is a vms layered
    > product that does this.


    Absolutely not. Plus it would cost extra, and it would perform
    much worse than the bound volume set.

  6. Re: Volume sets - yes, they can be very useful.

    "R.A.Omond" writes:

    >Phillip Helbig---remove CLOTHES to reply wrote:
    >> [...snip...]
    >>
    >> That should be "shadow set", not "volume set". A volume set is a
    >> different beast. (Yes, volume sets can be combined with shadow sets.
    >> In general, in these days of big disks, no-one needs volume sets
    >> anymore.)


    >Maybe in general, you're right. But at one of my customers,
    >we're into volume sets in a big way.


    >Like, 15 x 300 Gbyte disk volume sets.


    Since the loss of a single spindle can take out an entire volume set, be
    sure you are religious about your backups and make all the members
    shadowsets or redundant raidsets of some sort.

  7. Re: Clustering - how can I.......

    On Sat, 17 May 2008 05:58:47 -0700, Bob Gezelter
    wrote:

    > - a router that supported multiple external addresses each mapping to
    > a different internal IP address
    > - DNS entries that distribute the translation results over the
    > different external addresses.


    I would recomment getting a router with two NICs, e.g. Cisco 2621, whichis
    what I use, one for routable IP the others for internal then with 2 NICs
    in
    each VMS server, each going to a separate (managed) switch, Cisco 2924XLI
    have
    full redundancy. I run failsafe and lbroker and WASD on all of the
    servers.

    FREJA> tcpip ifconfig ie0
    IE0: flags=c43
    Standby addresses:
    inet 208.85.204.75 netmask ffffff00 broadcast 208.85.204.79
    (Active on F
    REJA WE0)
    *inet 10.0.0.11 netmask ff000000 broadcast 10.255.255.255 ipmtu 1500

    FREJA> tcpip ifconfig we0
    WE0: flags=c43
    Standby addresses:
    inet 10.0.0.11 netmask ff000000 broadcast 10.255.255.255 (Active
    on FREJ
    A IE0)
    *inet 208.85.204.75 netmask ffffff00 broadcast 208.85.204.79 ipmtu 1500

    I bought all my gear off ebay, and it wasn't expensive.





    --
    PL/I for OpenVMS
    www.kednos.com

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2