The SAN Storage Trinity

The SAN Storage Trinity
The SAN Storage Trinity
Posted on October 13, 2017 in Blog, Storage Blog
The following two tabs change content below.

Admin

Latest posts by Admin (see all)

​With Storage, the same design challenge always applies – balancing

  • Point 1: Terabyte Capacity
  • Point 2: IOP’s Performance – meaning true Random 4KB 75% Read / 25% Write ‘type’ IOPs
  • and Point 3: Cost

The above three form the holy trinity of a ‘basic’ Storage Design

With Storage, the same challenges always apply – balancing
    Terabyte Capacity
    IOP’s Performance – meaning true Random 4KB 75% Read / 25% Write ‘type’ IOPs
    and Cost
The above three form the holy trinity of a ‘base’ Storage Design

Tailored SAN Design


But now lets ​add other critical aspects which we need to tailor the storage to each customers taste.

Tailored SAN Design
But now lets throw in the essentials – that which we need/must add to the above to tailor the storage to each customers taste

​​Overall System Availability

  • Do we need a Single Controller / Single Node or Multiple Controllers to handle the failure of a single controller
  • In essence 99% of storage proposed is dual controller, unless we are talking about archive systems that could afford to be down for a day or so
OverAll System Availability
    Single Controller / Single Node or Multiple Controllers to handle the failure of a single controller
    In essence 99% of storage proposed is dual controller, unless we are talking about archive systems that could afford to be down for a day or so
With Storage, the same challenges always apply – balancing
    Terabyte Capacity
    IOP’s Performance – meaning true Random 4KB 75% Read / 25% Write ‘type’ IOPs
    and Cost
The above three form the holy trinity of a ‘base’ Storage Design

​  Data/Disk Redundancy

  • ​RAID 1/10, RAID5 or RAID 60
  • But once we make this selection, we immediately need to revisits Point 2
    • Our RAID Selection will critically impact our Write performance
    • and in RAID 10, it will at least halve our overall capacity – so we need to revisit Point 1
    • and now we need to Revisit Point 3
RAID 1/10, RAID5 or RAID 60
Ooops, but once we make this selection, we immediately need to revisits Point 2
    Our RAID Selection will critically impact our Write performance
    and in RAID 10, it will at least halve our overall capacity – so we need to revisit Point 1
    and now we need to Revisit Point 3
Data/Disk Redundancy
    RAID 1/10, RAID5 or RAID 60
    Ooops, but once we make this selection, we immediately need to revisits Point 2
        Our RAID Selection will critically impact our Write performance
        and in RAID 10, it will at least halve our overall capacity – so we need to revisit Point 1
        and now we need to Revisit Point 3
Tailored SAN Design
But now lets throw in the essentials – that which we need/must add to the above to tailor the storage to each customers taste

Compromise

At this point, we ​either have to:

  • ​add more Disks to increase capacity and IOPs
    • moved to SSD Disks for Performance
    • or taken another approach of adding a larger number of 7.2K RPM disks to achieve the same level of IOPS
Compromise
At this point, its likely we have either
    Had to add more Disks to increase capacity and IOPs either by
        and possibly
Compromise
At this point, its likely we have either
    Had to add more Disks to increase capacity and IOPs either by
        and possibly we’ve done this by using Faster Disks (eg if our design was based on 10K RPM Disks, we may have:
            moved to 15K RPM Disks for Performance
            or taken another approach of adding a larger number of 7.2K RPM disks to achieve the same level of IOPS
With Storage, the same challenges always apply – balancing
    Terabyte Capacity
    IOP’s Performance – meaning true Random 4KB 75% Read / 25% Write ‘type’ IOPs
    and Cost
The above three form the holy trinity of a ‘base’ Storage Design

​​​RAID

Introducing RAID, we have likely satisfied our ‘need’ to handle failure of 1, 2 or more disks

  • But a failed disk in a RAID needs to be re-built and that rebuild time can grow exponentially
  • ​Plan for 60-70-80 hours for a Rebuild of a 6TB drive in a RAID 6 array
RAID
Introducing RAID, we have likely satisfied our ‘need’ to handle failure of 1, 2 or more disks
    But a failed disk in a RAID needs to be re-built and that rebuild time can grow exponentially
        think 60-70-80 hours for a Rebuild of a 3TB drive in a RAID 6 array
Tailored SAN Design
But now lets throw in the essentials – that which we need/must add to the above to tailor the storage to each customers taste

​RAID Rebuild Times

Now we have introduced a new problem

  • Its possible the RAID rebuild time – and having our data at risk during this rebuild process does not meet our needs
  • OK – Lets go back and revisit RAID again – and maybe RAID 50, or RAID60 could alleviate some of the statistical problems associated with disk failure (ie spread, but not eliminate the risk)
RAID Rebuild Times
Now we have introduced a new problem
    Its possible the RAID rebuild time – and having our data at risk during this rebuild process does not meet our needs
    OK – Lets go back and revisit RAID again – and maybe RAID 50, or RAID60 could alleviate some of the statistical problems associated with disk failure (ie spread, but not eliminate the risk)
Compromise
At this point, its likely we have either
    Had to add more Disks to increase capacity and IOPs either by
        and possibly we’ve done this by using Faster Disks (eg if our design was based on 10K RPM Disks, we may have:
            moved to 15K RPM Disks for Performance
            or taken another approach of adding a larger number of 7.2K RPM disks to achieve the same level of IOPS
With Storage, the same challenges always apply – balancing
    Terabyte Capacity
    IOP’s Performance – meaning true Random 4KB 75% Read / 25% Write ‘type’ IOPs
    and Cost
The above three form the holy trinity of a ‘base’ Storage Design
Rackspace
While all of this is going on, the number of disks in our system has likely grown and grown
    Now we have a new problem
    We’re quickly running out of rackspace
        (let alone power or heat issues)
    Storage design was never meant to be easy – its an iterative process, where each change has a knock on effect to each an every other parameter.
Most small businesses visit this challenge once every 4 years.
Its an impossible decision to make and we often see over spend anywhere in the order of 30-40%.
This is justified by: ‘I can always use plenty of extra capacity’.
My take: You’ve likely burnt 30K GBP, 50K GBP , 100K or 200K GBP more than you needed to.
That spend could be put to improving your security, network or data protection position.
If you’re in the middle of a storage refresh and the deciding factor is how much discount you are about to receive, I hope you can pause for a moment and consider the points above.
RAID
Introducing RAID, we have likely satisfied our ‘need’ to handle failure of 1, 2 or more disks
    But a failed disk in a RAID needs to be re-built and that rebuild time can grow exponentially
        think 60-70-80 hours for a Rebuild of a 3TB drive in a RAID 6 array
Tailored SAN Design
But now lets throw in the essentials – that which we need/must add to the above to tailor the storage to each customers taste

​Rackspace and Power Costs


While all of this is going on, the number of disks in our system has likely grown and grown

  •  Now we have a new problem
  •  We’re quickly running out of rackspace
    •  (let alone power or heat issues)

    Storage design was never meant to be easy – its an iterative process, where each change has a knock on effect to each an every other parameter.

Rackspace
While all of this is going on, the number of disks in our system has likely grown and grown
    Now we have a new problem
    We’re quickly running out of rackspace
        (let alone power or heat issues)
    Storage design was never meant to be easy – its an iterative process, where each change has a knock on effect to each an every other parameter.

​In Closing

​Most small businesses visit this challenge once every 4 years.

Its an impossible decision to make and we often see over spend anywhere in the order of 30-40%.

​This is justified by: ‘I can always use plenty of extra capacity’.

My take: You’ve likely burnt 30K GBP, 50K GBP , 100K or 200K GBP more than you needed to.

That spend could be put to improving your security, network or data protection position.

If you’re in the middle of a storage refresh and the deciding factor is how much discount you are about to receive, I hope the points above offer some guidance.

Tags:, , , , , , ,