Storage and IOPS

Hi Everyone, This time I am going to share the most interesting article which is Storage. I heard many times from people  why storage is an important segment in IT infrastructure. My answer is, Storage playing a very critical role while designing  any IT infrastructure. Specially now days when technology is moving towards Virtual environment then storage consideration became more criticle and complex. In this article I would try to explain all those things and start explain everything from basics.

I would start each information in form of question and answer all those questions. This would clarify all your doubts while reading this article.

Q:What is storage?

Ans:Storage is a unit where all our data resides in form of bits/Bytes/KB/MB.GM/TB etc. Starting from OS related stuff to your personal data all goes to storage only.In ATA and SATA hard disks there are multiple magnetic storage disks bounded with free to rotate spindles, While coping or reading data from disks these spindles start rotating and provide required information, ATA spindle rotate at 7200RPM and SATA rotates at 15000RPM. 

Q:How many type storage available in market?

Ans:Specially storage is available  in three forms ATA/SATA/SSD. Each hard disk it's own specification based upon IOPS, The more IOPS capable HDD is the more cost would be.

Q:What is IOPS and why it's Important?

Ans:IOPS is input and outputs committed over the storage per second. Specially this defines the performance of storage and based upon that we decide which storage is best for us. The faster the Write rate and Read rate over the HDD the higher performance rate rate would be. Let me summerise some facts using which you can evaluate the performance of the disk,

– at the disk itself in Input-Output Per Second (IOPS) – a measure of how many read and write commands a disk can complete in a second

– interface throughput, measured in Mbps or Gbps – a measure of the peak rate that a volume of data can be read from or written to disk

 – latency – the amount of time between when you ask a disk (or storage system if you want to read ahead) to do something and when it can actually do it, very closely related to IOPS as you’ll read in a forthcoming article in this series. 

Q:How can we calculate the Disk IO?

Ans:We can use below formula to calculate the IO in any disk,

IOPS = 1000/(Seek Latency + Rotational Latency)

*1000=In one second we have 1000 milliseconds.

 *Seek Latency & Rotation Latency= This is mentioned over the disk so you can easy find it from there

Example:

– Average (rotational) latency: 2.0msec

– Average read seek (latency): 3.4msec

 – Average write seek (latency): 3.9msec

1000/2.0+3.4=185Maximum Read IOPS

Q:How can we check current IOPS rate at any server?

Ans:For this we can use Perfmon. In perfmon select

‘Disk Transfers/sec’ counter under Physical Disk.  This is a sum of the ‘Disk Reads/sec’ and ‘Disk Writes/sec’ counters as you can see in the screenshot below:

Q:How can we increase number of IOPS from storage?

Ans:To increase number of IOPS from storage we have to use multi disk structure. you can combine multiple hard drives together to aggregate their performance capabilities.  For example, two 15k RPM disks working together to server a workload could provide a theoretical 360 IOPS (180 + 180). This also  scales out so ten 15k RPM disks could provide 1800 IOPS, and 100 15k RPM disks could provide 18,000 IOPS.

Q:What is RAID?

Ans:RAID is short form of Redundant Array of Independent Disk. As it's self explanatory that to get more number of IOPS we need to use multiple disk's including redundancy and fault tolerance so here RAID concept come in to picture. 

RAID is implemented to increase performance through the aggregation of multiple disks, and to increase reliability though mirroring and parity.

We have 8 type of RAID configurations, To read more about RAID follow https://en.wikipedia.org/wiki/RAID . Incase you want me to write about it them please share your comments in below section and I would do it for you.

<<Drafting In Progress>>