This document provides a calculator to determine the number of disks needed for different RAID configurations based on frontend IOPS, read/write percentages, and disk type. The user enters the frontend IOPS and read/write percentage, and the calculator determines the backend IOPS for various RAID levels considering write penalties. It then calculates the number of disks needed based on the backend IOPS and average IOPS per disk type.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLS, PDF, TXT or read online on Scribd
0 ratings0% found this document useful (0 votes)
290 views2 pages
RAID Spindle Calculator
This document provides a calculator to determine the number of disks needed for different RAID configurations based on frontend IOPS, read/write percentages, and disk type. The user enters the frontend IOPS and read/write percentage, and the calculator determines the backend IOPS for various RAID levels considering write penalties. It then calculates the number of disks needed based on the backend IOPS and average IOPS per disk type.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLS, PDF, TXT or read online on Scribd
You are on page 1/ 2
RAID Spindle Calculator
Instructions:
Upper field: Enter IOs per seco
Frontend IOPS: 1000 Lower field: Enter percentage o enter values here calculated automatically) % Read IOs 67 % Write IOs 33 Caculation: The colums in yellow, orange and
Backend IOs calculated on each
RAID Configuration RAID 1 & 10 RAID 5 RAID 6 the given Read/Write distribution. random access pattern (database Backend IOPS (random): 1330 1990 2650 Number of Disks specifys the am RAID level with different disk type Number of Disks needed: typical average for random acces FC disks 180 IOPS 8 12 15 SAS Disk 200 IOPS 8 10 14 Legend: SATA Disk 75 IOPS 18 27 36 Frontend IOs: Total number of R application (e.g. MS Exchange) g
Backend IOs: Total number of R
RAID controller sends to its phys due to the fact that depending on perform multiple operations. Application Example:
1000 Frontend IOs; 67% Reads
(RAID 5 write penalty = 4; i.e. eac Backend. Each Read IO causes 1 Frontend IOs (between Application and Logical Volume) Reads: 1000 * 67% = 670 IOs Writes: 1000 * 33% = 330 IOs * 4 Total: 1990 IOs LUN Disks: 1990 IOs total / 200 per d
Backend IOs (between RAID Controller and physical disks) uctions:
er field: Enter IOs per second the application generates in total.
er field: Enter percentage of Read IOs (percentage of Write IOs
ulated automatically)
ulation: colums in yellow, orange and pink show:
kend IOs calculated on each RAID levels Write Penalty in subject to
given Read/Write distribution. Note: This calculation applies for mostly om access pattern (databases, file&print services etc.).
mber of Disks specifys the amount of disks needed in the particular
D level with different disk types. Note: IOPS values of disk types are al average for random access.
end:
ntend IOs: Total number of Read/Write operations per second an
cation (e.g. MS Exchange) generates and sends to its disk.
kend IOs: Total number of Read/Write operations per second the
D controller sends to its physical disks. This number is typically higher to the fact that depending on the RAID level the controller needs to orm multiple operations.