- Community Home
- >
- Storage
- >
- Entry Storage Systems
- >
- Disk Enclosures
- >
- Re: V7100 capacity
Disk Enclosures
1753404
Members
6875
Online
108793
Solutions
Forums
Categories
Company
Local Language
юдл
back
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
юдл
back
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Blogs
Information
Community
Resources
Community Language
Language
Forums
Blogs
Go to solution
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО09-09-2002 08:13 AM
тАО09-09-2002 08:13 AM
Hello--
There is something about the redundancy of the VA7100 that I don't understand.
Here is output from "armdsp".
CAPACITY USAGE
Total Disk Enclosures:________________1
Redundancy Group:_____________________1
Total Disks:________________________15
Total Physical Size:________________1,001.358 GB
Allocated to Regular LUNs:__________374.302 GB
Allocated as Business Copies:_______53 GB
Used as Active Hot Spare:___________66.753 GB
Used for Redundancy:________________192.846 GB
Unallocated (Available for LUNs):___314.455 GB
374.3 GBs are allocated to LUNs, and only 192.8 GBs are allocated for redundancy. There is 314 GB available for LUNs. I assume that the VA has some data in RAID 1+0 and some in RAID 5dp. Is this so? If so, how can I confirm this?
Since the VA has so much unused space, shouldn't the it be using more of this available disk for RAID 1+0 redundancy?
cjw
There is something about the redundancy of the VA7100 that I don't understand.
Here is output from "armdsp".
CAPACITY USAGE
Total Disk Enclosures:________________1
Redundancy Group:_____________________1
Total Disks:________________________15
Total Physical Size:________________1,001.358 GB
Allocated to Regular LUNs:__________374.302 GB
Allocated as Business Copies:_______53 GB
Used as Active Hot Spare:___________66.753 GB
Used for Redundancy:________________192.846 GB
Unallocated (Available for LUNs):___314.455 GB
374.3 GBs are allocated to LUNs, and only 192.8 GBs are allocated for redundancy. There is 314 GB available for LUNs. I assume that the VA has some data in RAID 1+0 and some in RAID 5dp. Is this so? If so, how can I confirm this?
Since the VA has so much unused space, shouldn't the it be using more of this available disk for RAID 1+0 redundancy?
cjw
Solved! Go to Solution.
2 REPLIES 2
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО09-09-2002 08:23 AM
тАО09-09-2002 08:23 AM
Solution
If it's in AutoRAID mode, then it's using both RAID levels. Check via CommandView what RAID mode you're in.
The VA will keep about 80% of your data in RAID 5DP, and about 20% in RAID 0/1. It will not exceed the 20% of RAID 0/1 unless it deems it necessary (and you have the extra capacity), which is rare.
Note that the VA is a kind of caching device with multiple levels - the highest cache level is the RAM cache on the controllers. The next highest cache level is RAID 0/1, and the lowest is RAID 5DP.
As you write to the array, it will initially store the hot-write data in RAM cache (mirroring it to the other contoller.) When it destages to disk, it writes it as RAID 0/1. Later is it migrated to RAID 5DP.
As an interesting side-effect, notice how it has to be able to keep a LUN in BOTH RAID levels at the same time. Some data blocks will be RAID 5DP, others RAID 0/1.
The reason it's not keeping more in RAID 0/1 is to net you a reasonable amout of capacity. If you're more interested in speed than capacity, lock it into RAID 0/1 mode.
Good luck!
The VA will keep about 80% of your data in RAID 5DP, and about 20% in RAID 0/1. It will not exceed the 20% of RAID 0/1 unless it deems it necessary (and you have the extra capacity), which is rare.
Note that the VA is a kind of caching device with multiple levels - the highest cache level is the RAM cache on the controllers. The next highest cache level is RAID 0/1, and the lowest is RAID 5DP.
As you write to the array, it will initially store the hot-write data in RAM cache (mirroring it to the other contoller.) When it destages to disk, it writes it as RAID 0/1. Later is it migrated to RAID 5DP.
As an interesting side-effect, notice how it has to be able to keep a LUN in BOTH RAID levels at the same time. Some data blocks will be RAID 5DP, others RAID 0/1.
The reason it's not keeping more in RAID 0/1 is to net you a reasonable amout of capacity. If you're more interested in speed than capacity, lock it into RAID 0/1 mode.
Good luck!
No matter where you go, there you are.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО09-10-2002 06:26 AM
тАО09-10-2002 06:26 AM
Re: V7100 capacity
There is a lot on confusion on how the VA uses capacity. Capacity in the array can be discussed in numerous ways (which adds to the confusion).
- physical, total capacity of the disks included in the array
- available or total usable, capacity presented to the administrator as available to create LUNs (sum of Unallocated, allocated to Regular LUNs, and allocated to Business Copy LUNs)
- reserved, subset of available, assigned or reserved for use by LUNs. Command View calls this ???allocated??? (sum of Allocated to Regular LUNs plus Allocated to Business Copy LUNs).
- Allocated (not the Command View usage), subset of reserved capacity, capacity that has been written to, modulo 256K (the minimum space allocated by the array, called a Cluster)
- redundancy space, sum of Clusters used as RAID 1+0 mirror copy, or RAID 5DP parity
- free space, Clusters not allocated
- reserved free space, Clusters logically removed from the usable capacity calculations to guarantee dead-lock free operation. (RAID migration and rebuild)
First point; think of the above output from Command View as a guaranteed capacity. Only when the LUNs are completely used (all blocks written), will allocated equal the reserved. Only the computation of the available capacity assumes a RAID 1+0 to RAID 5DP distribution (about 10% RAID 1+0 and 90% RAID 5DP of the physical capacity ??? not usable capacity). Second point; the real distribution can be dynamic. It is dependent on the current workload and the past 24-hour history of each Cluster.
To discover the current distribution of RAID 1+0 and RAID 5DP allocated capacity, you must use the Command View armperf command and look at the RAID 1+0 and RAID 5DP allocations. This metric is dynamic and will change as the array changes.
When configured in AutoRAID mode, the array is free to allocated any Cluster of a LUN to either RAID 1+0 or RAID 5DP. However, no single Cluster is simultaneously in both RAID levels ??? it???s either RAID 1+0 or RAID 5DP. But, a single LUN can have Clusters of both RAID levels.
Recent versions of the firmware are considerably different than earlier releases of the VA and its predecessor, the Model 12H. A common miss-conception from these versions is that the array always writes to RAID 1+0. Not so, if the workload can benefit and the space is available, writes can go directly to RAID 5DP. As and example, sequential write workloads can benefit for RAID 5DP, thus are directed to RAID 5DP.
- physical, total capacity of the disks included in the array
- available or total usable, capacity presented to the administrator as available to create LUNs (sum of Unallocated, allocated to Regular LUNs, and allocated to Business Copy LUNs)
- reserved, subset of available, assigned or reserved for use by LUNs. Command View calls this ???allocated??? (sum of Allocated to Regular LUNs plus Allocated to Business Copy LUNs).
- Allocated (not the Command View usage), subset of reserved capacity, capacity that has been written to, modulo 256K (the minimum space allocated by the array, called a Cluster)
- redundancy space, sum of Clusters used as RAID 1+0 mirror copy, or RAID 5DP parity
- free space, Clusters not allocated
- reserved free space, Clusters logically removed from the usable capacity calculations to guarantee dead-lock free operation. (RAID migration and rebuild)
First point; think of the above output from Command View as a guaranteed capacity. Only when the LUNs are completely used (all blocks written), will allocated equal the reserved. Only the computation of the available capacity assumes a RAID 1+0 to RAID 5DP distribution (about 10% RAID 1+0 and 90% RAID 5DP of the physical capacity ??? not usable capacity). Second point; the real distribution can be dynamic. It is dependent on the current workload and the past 24-hour history of each Cluster.
To discover the current distribution of RAID 1+0 and RAID 5DP allocated capacity, you must use the Command View armperf command and look at the RAID 1+0 and RAID 5DP allocations. This metric is dynamic and will change as the array changes.
When configured in AutoRAID mode, the array is free to allocated any Cluster of a LUN to either RAID 1+0 or RAID 5DP. However, no single Cluster is simultaneously in both RAID levels ??? it???s either RAID 1+0 or RAID 5DP. But, a single LUN can have Clusters of both RAID levels.
Recent versions of the firmware are considerably different than earlier releases of the VA and its predecessor, the Model 12H. A common miss-conception from these versions is that the array always writes to RAID 1+0. Not so, if the workload can benefit and the space is available, writes can go directly to RAID 5DP. As and example, sequential write workloads can benefit for RAID 5DP, thus are directed to RAID 5DP.
The opinions expressed above are the personal opinions of the authors, not of Hewlett Packard Enterprise. By using this site, you accept the Terms of Use and Rules of Participation.
News and Events
Support
© Copyright 2024 Hewlett Packard Enterprise Development LP