Sunday, 23 September 2012

RAID Levels:

RAID (Redundancy Array Of Independent Disks): 




RAID 0:

Technology: Striping Data with No Data Protection.
Performance: Highest
Overhead: None
Minimum Number of Drives: 2 since striping
Data Loss: Upon one drive failure
Example: 5TB of usable space can be achieved through 5 x 1TB of disk.
Advantages:
>
High Performance
Disadvantages: Guaranteed Data loss
Hot Spare: Upon a drive failure, a hot spare can be invoked, but there will be no data to copy over. Hot Spare is not a good option for this RAID type.
Supported: Clariion, Symmetrix, Symmetrix DMX (Meta BCV’s or DRV’s)
In RAID 0, the data is written / stripped across all of the disks. This is great for performance, but if one disk fails, the data will be lost because since there is no protection of that data.


RAID 1:

Technology: Mirroring and Duplexing
Performance: Highest
Overhead: 50%
Minimum Number of Drives: 2
Data Loss: 1 Drive failure will cause no data loss. 2 drive failures, all the data is lost.
Example: 5TB of usable space can be achieved through 10 x 1TB of disk.
Advantages: Highest Performance, One of the safest.
Disadvantages: High Overhead, Additional overhead on the storage subsystem. Upon a drive failure it becomes RAID 0.

Hot Spare: A Hot Spare can be invoked and data can be copied over from the surviving paired drive using Disk copy.
Supported: Clariion, Symmetrix, Symmetrix DMX
The exact data is written to two disks at the same time. Upon a single drive failure, no data is lost, no degradation, performance or data integrity issues. One of the safest forms of RAID, but with high overhead. In the old days, all the Symmetrix supported RAID 1 and RAID S. Highly recommended for high end business critical applications.
The controller must be able to perform two concurrent separate Reads per mirrored pair or two duplicate Writes per mirrored pair. One Write or two Reads are possible per mirrored pair. Upon a drive failure only the failed disk needs to be replaced.


RAID 1+0 :

Technology: Mirroring and Striping Data
Performance: High
Overhead: 50%
Minimum Number of Drives: 4
Data Loss: Upon 1 drive failure (M1) device, no issues. With multiple drive failures in the stripe (M1) device, no issues. With failure of both the M1 and M2 data loss is certain.
Example: 5TB of usable space can be achieved through 10 x 1TB of disk.
Advantages: Similar Fault Tolerance to RAID 5, Because of striping high I/O is achievable.
Disadvantages: Upon a drive failure, it becomes RAID 0.
Hot Spare: Hot Spare is a good option with this RAID type, since with a failure the data can be copied over from the surviving paired device.
Supported: Clariion, Symmetrix, Symmetrix DMX
RAID 1+0 is implemented as a mirrored array whose segments are RAID 0 arrays.


RAID 3 :

Technology: Striping Data with dedicated Parity Drive.
Performance: High
Overhead: 33% Overhead with Parity (in the example above), more drives in Raid 3 configuration will bring overhead down.
Minimum Number of Drives: 3
Data Loss: Upon 1 drive failure, Parity will be used to rebuild data. Two drive failures in the same Raid group will cause data loss.
Example: 5TB of usable space would be achieved through 9 1TB disk.
Advantages: Very high Read data transfer rate. Very high Write data transfer rate. Disk failure has an insignificant impact on throughput. Low ratio of ECC (Parity) disks to data disks which converts to high efficiency.
Disadvantages: Transaction rate will be equal to the single Spindle speed
Hot Spare: A Hot Spare can be configured and invoked upon a drive failure which can be built from parity device. Upon drive replacement, hot spare can be used to rebuild the replaced drive.
Supported: Clariion


RAID 5 :

Technology: Striping Data with Distributed Parity, Block Interleaved Distributed Parity
Performance: Medium
Overhead: 20% in our example, with additional drives in the Raid group you can substantially bring down the overhead.
Minimum Number of Drives: 3
Data Loss: With one drive failure, no data loss, with multiple drive failures in the Raid group data loss will occur.
Example: For 5TB of usable space, we might need 6 x 1 TB drives
Advantages: It has the highest Read data transaction rate and with a medium write data transaction rate. A low ratio of ECC (Parity) disks to data disks which converts to high efficiency along with a good aggregate transfer rate.
Disadvantages: Disk failure has medium impact on throughput. It also has most complex controller design. Often difficult to rebuild in the event of a disk failure (as compared to RAID level 1) and individual block data transfer rate same as single disk. Ask the PSE’s about RAID 5 issues and data loss?
Hot Spare: Similar to RAID 3, where a Hot Spare can be configured and invoked upon a drive failure which can be built from parity device. Upon drive replacement, hot spare can be used to rebuild the replaced drive.
Supported: Clariion, Symmetrix DMX code 71
RAID Level 5 also relies on parity information to provide redundancy and fault tolerance using independent data disks with distributed parity blocks. Each entire data block is written onto a data disk; parity for blocks in the same rank is generated on Writes, recorded in a distributed location and checked on Reads.
This would classify to be the most favorite RAID Technology used today.



RAID 6 :

Technology: Striping Data with Double Parity, Independent Data Disk with Double Parity
Performance: Medium
Overhead: 28% in our example, with additional drives you can bring down the overhead.
Minimum Number of Drives: 4
Data Loss: With one drive failure and two drive failures in the same Raid Group no data loss. Very reliable.
Example: For 5 TB of usable space, we might need 7 x 1TB drives
Advantages: RAID 6 is essentially an extension of RAID level 5 which allows for additional fault tolerance by using a second independent distributed parity scheme (two-dimensional parity). Data is striped on a block level across a set of drives, just like in RAID 5, and a second set of parity is calculated and written across all the drives; RAID 6 provides for an extremely high data fault tolerance and can sustain multiple simultaneous drive failures which typically makes it a perfect solution for mission critical applications.
Disadvantages: Very poor Write performance in addition to requiring N+2 drives to implement because of two-dimensional parity scheme.
Hot Spare: Hot Spare can be invoked against a drive failure, built it from parity or data drives and then upon drive replacement use that hot spare to build the replaced drive.
Supported: Clariion Flare 26, 28, Symmetrix DMX Code 72, 73
Clariion Flare Code 26 supports RAID 6. It is also being implemented with the 72 code on the Symmetrix DMX. The simplest explanation of RAID 6 is double the parity. This allows a RAID 6 RAID Groups to be able to have two drive failures in the RAID Group, while maintaining access to the data.

RAID S (3+1) :

Technology: RAID Symmetrix
Performance:  > High
Overhead: 25%
Minimum Number of Drives: 4
Data Loss: Upon two drive failures in the same Raid Group
Example: For 5 TB of usable space, 8 x 1 TB drives
Advantages: High Performance on Symmetrix Environment
Disadvantages: Proprietary to EMC. RAID S can be implemented on Symmetrix 8000, 5000 and 3000 Series. Known to have backend issues with director replacements, SCSI Chip replacements and backend DA replacements causing DU or offline procedures.
Hot Spare: Hot Spare can be invoked against a failed drive, data can be built from the parity or the data drives and upon a successful drive replacement, the hot spare can be used to rebuild the replaced drive.
Supported: Symmetrix 8000, 5000, 3000. With the DMX platform it is just called RAID (3+1)
EMC Symmetrix / DMX disk arrays use an alternate, proprietary method for parity RAID that they call RAID-S. Three Data Drives (X) along with One Parity device. RAID-S is proprietary to EMC but seems to be similar to RAID-5 with some performance enhancements as well as the enhancements that come from having a high-speed disk cache on the disk array.
The data protection feature is based on a Parity RAID (3+1) volume configuration (three data volumes to one parity volume).

RAID (7+1):

Technology: RAID Symmetrix
Performance: High
Overhead: 12.5%
Minimum Number of Drives: 8
Data Loss: Upon two drive failures in the same Raid Group
Example: For 5 TB of usable space, 8 x 1 TB drives (rather you will get 7 TB)
Advantages: High Performance on Symmetrix Environment
Disadvantages: Proprietary to EMC. Available only on Symmetrix DMX Series. Known to have a lot of backend issues with director replacements, backend DA replacements since you have to verify the spindle locations. Cause of concern with DU.
Hot Spare: Hot Spare can be invoked against a failed drive, data can be built from the parity or the data drives and upon a successful drive replacement, the hot spare can be used to rebuild the replaced drive.
Supported: With the DMX platform it is just called RAID (7+1). Not supported on the Symms.
EMC DMX disk arrays use an alternate, proprietary method for parity RAID that is called RAID. Seven Data Drives (X) along with One Parity device. RAID is proprietary to EMC but seems to be similar to RAID-S or RAID5 with some performance enhancements as well as the enhancements that come from having a high-speed disk cache on the disk array.
The data protection feature is based on a Parity RAID (7+1) volume configuration (seven data volumes to one parity volume).

What is Vios?

What is Vios? 


 The virtual i/o server is an appliance that provides virtual storage and shared ethernet adapter capability to client logical partitions.it allow a physical adapter with attached disks on the virtual i/o sever partition to be shared by one or more partitions,enbling clients to consolidate and potentially minimize the number of physical adapters required.


VIOS is a special purpose partition that can serve I/O resources to other partitions. The type of LPAR is set at creation. The VIOS LPAR type allows for the creation of virtual server adapters, where a regular AIX/Linux LPAR does not.
• VIOS works by owning a physical resource and mapping that physical resource to virtual resources. Client LPARs can connect to the physical resource via these mappings.
• VIOS is not a hypervisor, nor is it required for sub-CPU virtualization. VIOS can be used to manage other partitions in some situations when a HMC is not used. This is called IVM (Integrated Virtualization Manager).


Depending on configurations, VIOS may or may not be a single point of failure. When client partitions access I/O via a single path that is delivered via in a single VIOS, then that VIOS represents a potential single point of failure for that client partition.

• VIOS is typically configured in pairs along with various multipathing / failover methods in the client for virtual resources to prevent the VIOS from becoming a single point of failure.
• Active memory sharing and partition mobility require a VIOS partition. The VIOS partition acts as the controlling device for backing store for active memory sharing. All I/O to a partition capable of partition mobility must be handled by VIOS as well as the process of shipping memory between physical systems.



Tuesday, 4 September 2012

HMC Basic Tips:

HMC Basic Questions :


1. What is the maximum number of servers managed by HMC ?

- Maximum of 48 non-590-595 servers
- Maximum of 32 590/595 servers

2. What is the maximum number of LPARs supported by a HMC ?

- Maximum of 254 LPARs

3. How many HMCs can manage a server at one time ?

- You can have a maximum of 2 HMCs manage a server at one time

4. What are the different types of dynamic operations you can do with CPU, Memory and I/O Adapter on a LPAR ?

-Add
- Remove
- Move

5. How do we connect the HMC to power machines ?

For Power-4 machines, we connect the hmc using serial cables.

But for Power-5 machines, HMC connects to service processors via SSL-encrypted Ethernet, replacing the serial cables.

6. Do we have firewall configured in HMC ?

Yes. Each network card has an integrated firewall.

7. Do we need to configure DHCP in HMC ?

HMC may be a DHCP server for entry and mid-range servers.

But for high-end servers like P595, HMC must be a DHCP server

8. can we have the same HMC to manage P4 and P5 machines ?

POWER5 HMCs cannot manage POWER4 servers, and vice versa.

9. Can we have the existing P4 HMCs upgraded to support P5 machines ?

Yes. We can. This involves a complete overwirte of teh disk and the loss of all previous configuration including user profiles.

10. What to do incase of disk failure in HMC ?

We can restore the HMC using recovery CD.
Then restore the latest Critical consule data backup which will restore the profiles, user ids, passwords, etc..

11. What is the default user id and password for the HMC ?

When the HMC is powered on the first time, login as hscroot and password as 'abc123'.

12. Can we manage a power machine without a HMC ?

Yes. We can run a server in manufacturing default mode, will all resources but no logical partitionings, CoD or Service Focal point,etc..
For entry level server, we can use the Integrated Virtualization Manager.

13. What is the network critetia for dual HMC connection ?

Dual HMCs require two different private networks.

14. What is the default service processor IP address in Power-5 Machines ?

Eth0 - HMC1 - 169.254.2.147 / 255.255.255.0
Eth1 - HMC2 - 169.254.3.147 / 255.255.255.0

15. What is the default user id and password for accessing service processor ?

User id - admin
Password - admin

16. Do we need a HMC for p5 model servers ?

One HMC is mandatory for 590, 595 or 575.
Dual HMC are recommended.

17. Do we need private network for HMc connectivity for p5-595 ?

One private network is mandatory for p5 590,595 or 575.

18. Can we have IVM support multiple servers ?

One IVM allowed per server and it only manages partitions on one server.

19. What does FSP (Flexible Service Processor) has ?

FSP has
a. Operating System
b. UserIds / Passwords
c. Filesystem
d. Networking
e. Firewall
f. Webserver
g. ASMI
h. Firmware

20. What to do if you forgot the admin password for FSP ?

If you do not know the admin password, place a hardware call to get 'celogin'

21. What to do if you forgot the HMC hostname/ipaddress for a long running LPAR ?

You can always get the HMC IPaddress from a LPAR if we have performed "handshake" with the HMC.

Issue the below command to get the HMC IPAddress

# lsrsrc IBM.ManagementServer

Resource Persistent Attributes for IBM.ManagementServer
resource 1:
Name = "169.121.54.48"
Hostname = "169.121.54.48"
ManagerType = "HMC"
LocalHostname = "169.121.54.59"
ClusterTM = "9078-160"
ClusterSNum = ""
ActivePeerDomain = ""
NodeNameList = {"SAP-PRodServer"}

22. One HMC should be within 8metres of Managed Server

23. Each FSP Ethernet port should be connected to only one HMC

Paging Space Concept In Aix.


What Is Paging Space?
Paging space is a virtual memory which is an illusion to OS as if it had more real memory.


# lsps -a -To all all paging spaces

Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv

CREATE A NEW LV TO USE FOR PAGING SPACE
 :


# mklv -t paging rootvg 5
pagelv00


SEE CURRENT PAGING SPACES WHICH ARE  REGISTERED :

#
# cat /etc/swapspaces
* /etc/swapspaces
*
* This file lists all the paging spaces that are automatically put into
* service on each system restart (the 'swapon -a' command executed from
* /etc/rc swaps on every device listed here).
*
* WARNING: Only paging space devices should be listed here.
*
* This file is modified by the chps, mkps and rmps commands and referenced
* by the lsps and swapon commands.
hd6:
        dev = /dev/hd6


ENABLING A PAGING SPACE :


# swapon /dev/pagelv00
# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
pagelv00        hdisk0            rootvg         640MB     1   yes    no    lv
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv
#


DISABLING A PAGING SPACE :


# swapoff /dev/pagelv00
# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
pagelv00        hdisk0            rootvg         640MB     0    no    no    lv
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv
#


CREATING A PAGING SPACE THROUGH mkps :


# mkps -a -n -s3 rootvg
paging00
# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
paging00        hdisk0            rootvg         384MB     1   yes   yes    lv
pagelv00        hdisk0            rootvg         640MB     0    no    no    lv
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv
MKPS AUTOMATICALLY ADDS PAGING SPACE IN /etc/swapspaces 


# cat /etc/swapspaces
* /etc/swapspaces
*
* This file lists all the paging spaces that are automatically put into
* service on each system restart (the 'swapon -a' command executed from
* /etc/rc swaps on every device listed here).
*
* WARNING: Only paging space devices should be listed here.
*
* This file is modified by the chps, mkps and rmps commands and referenced
* by the lsps and swapon commands.
hd6:
        dev = /dev/hd6


paging00:
        dev = /dev/paging00


CHANGING PAGING SPACE :


INCREASING PAGING SIZE :



# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
paging00        hdisk0            rootvg         384MB     1   yes   yes    lv
pagelv00        hdisk0            rootvg         640MB     0    no    no    lv
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv


# chps -s 2 paging00


# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
paging00        hdisk0            rootvg         640MB     1   yes   yes    lv
pagelv00        hdisk0            rootvg         640MB     0    no    no    lv
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv
#


DECREASING PAGING SPACE :



# chps -d 1 paging00
shrinkps: Temporary paging space paging01 created.
shrinkps: Paging space paging00 removed.
shrinkps: Paging space paging00 recreated with new size.



# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
paging00        hdisk0            rootvg         512MB     1   yes   yes    lv
pagelv00        hdisk0            rootvg         640MB     0    no    no    lv
hd6             hdisk0            rootvg         512MB    17   yes   yes    lv
#