For many year storage wasn’t seen as important as processing power and powerful applications. One of the reasons is there wasn’t much you could do to improve storage performance. With recent innovations in solid sate disks (SSD), this has changed to make your ability to improve disk performance easier. We can now improve your data center infrastructure by removing old hard disk (HDD) components and creating a data center that takes up just a small a fraction of the space than previous generations.
What are the 5 things you can do to build a more efficient data center?
1. Understand Your Workload
One of the biggest mistakes you can make is to build a solution based on wrong assumptions. One of the most commonly made mistakes is to build with excessive margin based on incorrect workload assumptions. In a design built around traditional HDD, we may add more spindles than needed to cover incorrect IOPS requirements, thus wasting money to solve capacity issues. With a SSD solution, make sure to understand the balance of performance and write endurance. Actual workloads may surprise you. Spending time to really understand the workload can save you a lot of money.
2. Understand the “Access Density” Needed
This is often referred to in terms of IOs per second per GB. This statistic can mean the difference between overworking a HDD, using an all SSD, or using direct attached flash in the server. It can also change the type of protection a solution uses (RAID 5, RAID 10, etc.) or the type of SSD chosen.
The ratio of controllers versus capacity coupled with the parallelism that comes with it is the single most important access density point to remember when deciding what to build and why.
3. Hardware Consolidation
A few large capacity SSDs to help consolidate datacenter hardware, potentially saving you money and space. Comparing cost per GB of devices is probably too simplistic. Shrinking the hardware footprint allows for removal of enclosures and other support infrastructure which changes the “solution cost per GB” to a larger extent than device comparisons will illustrate. While SSD will usually be more expensive than HDD, you can probably buy fewer drive overall, and this will save you money and reduce your hardware failure footprint.
4. Simple Data Center
You want fewer, components that do exactly what you need them to do, which will allow you to save money on initial hardware cost and ongoing maintenance. This could allow for a complete reimagining of the data center. Use components that allow for simpler solutions. Clearly, the SSD solution provides the capacity/performance needed at a similar or small incremental price, but it could result in operating cost savings.
5. Understand Changes
Question everything. Technology changes and what you know as standard configuration today may be old news tomorrow. Cost, capacity, level of protection, and performance will change and will impact what types of systems you will build. A RAID 10 HDD solution used in the past (for speed and reliability) may be replaced with SSD using RAID 5 today. An application may have no use for SSD levels of performance but can benefit greatly from it during failure scenarios.
Spending time understanding your options may save you a few headaches later.