Jump to content

jcovey

Members
  • Posts

    3
  • Joined

  • Last visited

Posts posted by jcovey

  1. Hey everybody!  I have heard several times that Intella is being used for multi-terabyte cases, which I don't at all doubt.  What I'd like to know is, for those doing so, what is your specific disk setup?  For example, I have a high-end Dell Precision rackmount workstation on which I run Intella Pro.  It has 5 physical disks with RAID 5.  It's basically ~ 300 GB system drive, 1 TB each for case data and case index, and 500 GB for Optimization.  I have 3 empty slots to add more drives, but with 10K RPM SAS drives, the disks are only 900 GB.  So even if we max it out with 3 more, I still doubt that I can achieve, for example, a 2.5 TB case data, case index, and a 1+ GB Optimization drive if I retain RAID.  I'm assuming most everyone is using RAID?  I certainly don't want the reduced performance of larger, 7200 RPM drives, so what to do?

     

    We also use Intella Connect, and have two Intella Node instances.  This lets us process data on a production SAN with fiber channel connection.  If I want, I can add fiber channel to my Dell in order to give it high bandwidth SAN access.  We're also just getting a new Tintri, hybrid SAN online.  Intella Node processing has become so fast on the SAN that I'm not even sure that the Dell machine can edge it out any more, and the hybrid system will certainly be faster.  Regardless, these are my other disk options.  

     

    The problem we run into has to do with backups.  We use Veeam and are fully virtualized, and our network engineer tells me that having a lun larger than 2 TB causes issues with Veeam backups and/or VMWare. Yet I know some of you must be dealing with this, so I thought I would go straight to the source and ask how!  Any info would be greatly appreciated.  Thanks!  

     

    Jason    

  2. So the questions I have all revolve around the Remote Processors concept and how exactly they will function and where the data will reside.  I get the concept of centralized control of your Intella machines via the expanded web interface.  However, with all processing performed locally on processing machines, where does the data reside for access by Connect once processing is completed?  I'm specifically envisioning my high end workstation, which is dedicated to processing, after which I manually move the data to centralized storage for access via VMs.  Will that still be the case?

     

    A major new issues emerged for me in that, despite my new machine, when I inevitably have to add add'l data, I'm having to manually move a 200 GB database BACK to that machine, process, then manually move it AGAIN to the centralized storage (SAN-based in my case).  

     

    Could you perhaps clarify how these new capabilities will work from that perspective?  Thanks!

     

    Jason   

×
×
  • Create New...