This post is more than 5 years old
1 Rookie
•
10 Posts
0
6698
March 2nd, 2016 13:00
Can somebody explain how capacity works in ScaleIO?
This is my capacity. I have 9 drives of ~2TB each. I understand that 16.4 is about what it comes out to. In a raid 1 situation, I see I have roughly 7.8 TB of usable space. Can someone explain what the 3TB of unused space is there?
Also, I only have 2 2TB volumes created and mapped to SDCs, Why is it that I only have 1.5TB Free?
Thanks!
torage Pool SP1 (Id: xxxxxxxxxx) has 2 volumes and 1.5 TB (1520 GB) available for volume allocation
The number of parallel rebuild/rebalance jobs: 2
Rebuild is enabled and using Limit-Concurrent-IO policy with the following parameters:
Number of concurrent IOs per device: 1
Rebalance is enabled and using Favor-Application-IO policy with the following parameters:
Number of concurrent IOs per device: 1, Bandwidth limit per device: 10240 KB per second
Background device scanner: Disabled
Zero padding is enabled
Spare policy: 34% out of total
Uses RAM Read Cache
RAM Read Cache write handling mode is 'cached'
16.4 TB (16754 GB) total capacity
3.0 TB (3057 GB) unused capacity
0 Bytes snapshots capacity
7.8 TB (8000 GB) in-use capacity
0 Bytes thin capacity
7.8 TB (8000 GB) protected capacity
0 Bytes failed capacity
0 Bytes degraded-failed capacity
0 Bytes degraded-healthy capacity
0 Bytes unreachable-unused capacity
0 Bytes active rebalance capacity
0 Bytes pending rebalance capacity
0 Bytes active fwd-rebuild capacity
0 Bytes pending fwd-rebuild capacity
0 Bytes active bck-rebuild capacity
0 Bytes pending bck-rebuild capacity
0 Bytes rebalance capacity
0 Bytes fwd-rebuild capacity
0 Bytes bck-rebuild capacity
0 Bytes active moving capacity
0 Bytes pending moving capacity
0 Bytes total moving capacity
5.6 TB (5696 GB) spare capacity
7.8 TB (8000 GB) at-rest capacity
0 Bytes decreased capacity
Primary-reads 1 IOPS 1.6 KB (1638 Bytes) per-second
Primary-writes 40 IOPS 238.6 KB (244326 Bytes) per-second
Secondary-reads 0 IOPS 0 Bytes per-second
Secondary-writes 43 IOPS 232.0 KB (237568 Bytes) per-second
Backward-rebuild-reads 0 IOPS 0 Bytes per-second
Backward-rebuild-writes 0 IOPS 0 Bytes per-second
Forward-rebuild-reads 0 IOPS 0 Bytes per-second
Forward-rebuild-writes 0 IOPS 0 Bytes per-second
Rebalance-reads 0 IOPS 0 Bytes per-second
Rebalance-writes 0 IOPS 0 Bytes per-second
Volumes summary:
2 thick-provisioned volumes. Total size: 3.9 TB (4000 GB)
tomer__engineer
155 Posts
0
March 3rd, 2016 00:00
How many SP do you have?
How many SDSs do you have? 3? and in each SDS 3 drives?
As you can see in the output the spare is configured to 34% (rounded up from 33.33334).
Spare is calculated to be able to protect "loosing" 1 fault unit, if they are all even (Basically, your largest fault unit).
Fault unit can be a single SDS, or a Fault Set (FS) made out of several SDSs.
So in you case 16.4 TB * 34/100 = 5.6 TB spare capacity (Blue section)
You created 2 * ~2 TB thick volumes = ~4 TB (but due to the RAID1 it's actually ~8 TB (Green section)
So what is left is 3 TB available for volume allocation, but again, this needs to be divided to 2 (RAID1), so you have only 1.5 TB available for volume allocation (Grey section)
This is why you can have better disk space utilization to create a 30 TB storage made of 10 nodes/SDS (in that case if they are all even spare will be only 10%), rather than 3 nodes. And as a bonus, you will get better performance due to ScaleIO's parallel disk utilization across all SDSs.
allunr
1 Rookie
•
11 Posts
1
March 2nd, 2016 20:00
ScaleIO reserves 10% of capacity by default for data protection in case of unplanned downtime. It is reserved when rebuilds are going to require unused disk space.
CFrank2
1 Rookie
•
10 Posts
0
March 3rd, 2016 07:00
This helps! This is what I was looking for. I never could find a straight answer how things were laid out. I assumed the Green area (7.8TB) was my available space (after RAID) for volumes, not volumes I configured including RAID.
So since all of my SDS's have the same size drives in them, my spare capacity equals 1 SDS size.
Yes, I have a basic config with 3 servers, each with an SDS on them. Each server has 3 2TB (1.8TB) drives. This is in my basement lab, so I can't really spin up 10 nodes...the wife would kill me.
Good information. Thanks a ton!