WMarow's IOPS calculator
Deze tool kan je gebruiker voor het berekenen van de performance en capaciteit van hardeschijven, SSD's en RAID array's. Het gemiddelde random IOPS en de netto capaciteit van een RAID array wordt berekend op basis van de parameters van een schijf, het totaal aan schijven, type RAID en type belasting
De oorspronkelijke pagina van Marek Wołynko (wmarow) is offline gehaald toen hij stopte met IT gerelateerd werk. Deze pagina is daarna door Xaminmo (Omnitech Industries) uit web.archive.org gehaald en op zijn eigen domein gehost. Hier hebben wij weer deze kopie van gemaakt en beschikbaar gesteld voor algemeen nut.
Wij hebbenhet lijstje met sample disks aangepast met een set aan standaard HPE hardeschijven
wmarow's iops calculator
disk capacity (GB)
Every drive has a space for our data. The size of the drive is given in its specification
and its expressed in GB or TB.
1 GB (gigabyte) = 1000 * 1000 * 1000 bytes = 1000000000 bytes
close
disk rotational speed (RPM)
Working HDD has to spin. The faster it spins its performance is higher.
This parameter is expressed in number of complete revolutions per minute.
The fastest drives spin at a speed of 15,000 RPM.
close
average rotational latency (RPM)
The time required to complete half of the rotation is the average rotation latency.
This is the first of the two parameters that determine the performance of the disk.
The second one is average seek latency.
close
average disk seek latency (ms)
This parameter tells us how fast the disk head moves.
It's the average time it takes the head assembly on the actuator arm to
travel to the track of the disk where the data will be read or written. Read and write
operations have different seek latency. That's why we use the average value.
close
average drive operating power (W)
Working HDD and SSD needs power. In the drive's specification you can find what is the average power
consumption of operating drive.
close
SSD read iops
There is no formula to calculate SSD perforance. You have to believe in the information from the manufacturer or
perform reliable tests.
close
SSD write iops
There is no formula to calculate SSD perforance. You have to believe in the information from the manufacturer or
perform reliable tests.
close
RAID level
RAID, acronym for Redundant Array of Independent Disks
(originally Redundant Array of Inexpensive Disks),
is a technology that provides increased storage functions and reliability through redundancy.
This is achieved by combining multiple disk drive components into a logical unit,
where data is distributed across the drives in one of several ways called "RAID levels" -
Read more at Wikipedia
close
stripe size (KiB)
RAID arrays improve performance by splitting up files into small pieces
and distributing them to multiple hard disks. These small pieces are called stripes and they have size
- stripe size.
close
number of drives in the array
How many HDD or SSD you have?
close
drives used in the calculation
To meet the requirements listed above (raid level and it's length),
the array can have just as many drives.
close
reads (%)
Reading data or writing data - these two operations can execute applications on the disk array.
You can find read and write percentage in the statistics of operations on disk array, operating systems or applications.
close
read cache hit ratio (%)
In computer engineering, a cache is a component that transparently stores data
so that future requests for that data can be served faster.
The data that is stored within a cache might be values that have been computed earlier or
duplicates of original values that are stored elsewhere.
If requested data is contained in the cache (cache hit),
this request can be served by simply reading the cache, which is comparatively faster.
Otherwise (cache miss), the data has to be recomputed or fetched from its original storage location,
which is comparatively slower. Hence, the more requests can be served from the cache the
faster the overall system performance is. (source: Wikipedia)
Cache hit ratio can be found in statistics of the disk array.
close
write cache hit ratio (%)
When write caching is enabled, when the system sends a write to the hard disk,
the logic circuit records the write in its much faster cache, and then immediately
sends back an acknowledgement to the operating system saying, in essence, "all done!"
The rest of the system can then proceed on its merry way without having to sit around
waiting for the actuator to position and the disk to spin, and so on.
This is called write-back caching, because the data is stored in the cache and only "written back"
to the platters later on.(source: PC guide)
Cache hit ratio can be found in statistics of the disk array.
close
average I/O size (KiB)
Each read and write operation contains some data. Their size may vary.
It depends on the size of the block which operates an application,
as well as operating system settings. The average size of I/O operations
can be found in the statistics of disk array or operating system.
close
queue length and IOPS
Operating systems have the ability to set the length queue for IO operations.
This parameter determines the number of IO operations that can be sent to the
array without waiting for a response. Too low value for this parameter
can limit the number of possible IOPS. Too large value may exhaust the limit
of unconfirmed operations on array's port.
close
number of FC ports in the server
How many FC ports are in your server?
close
average I/O service time (ms)
How fast your array responds?
close
FC port queue depth
It's OS parameter.
close
lun queue depth
It's OS parameter.
close
disk capacity (GiB)
Disk capacity presented by the manufacturers of these devices is expressed in GB or TB.
These units are created by increasing the power of the number 1000.
Meanwhile, the computers shows the size of disks in GiB (gibibyte) or TIB (tebibyte).
They are formed by increasing the power of the number 1024.
1 GiB (gibibyte) = 1024 * 1024 * 1024 bytes = 1073741824 bytes
For the binary system it's more natural. Read more at Wikipedia
close
average random disk IOPS
The hard drive as opposed to tape is a random access device and in most cases in this way is used.
Therefore, the performance of this device should be determined in average
random Input / Output operations per second (IOPS).
This parameter is calculated based on values of the delays generated by the moving drive parts
- rotating disk and positioning the head.
close
disk access density (IOPS/GiB)
Access Density is the ratio of performance, measured in I/Os per second,
to the capacity of the drive
close
array native capacity (GiB)
Raw capacity of the array, without RAID protection, spare disks and so on.
close
array usable capacity (GiB)
Disk array capacity available for use
close
average random IOPS
This is not a simple sum of disk IOPS.
That's why you need to enter all these parameters on the left.
close
bandwidth (MiB/s)
Why so slow? It's a bandwidth of random operations. In the real world,
there are also sequential IOs, the bandwidth will be greater.
close
average power consumption of disks
The amount of energy consumed by all the drives in a disk array.
close
maximum host IOPS
Maximum IO performance for all FC ports.
close
maximum lun IOPS
Maximum IO performance for a single lun.
close