Powerpath Fundamentals - SRG PDF

Copyright 2013 EMC Corporation. Do not copy - All Rights Reserved.
Welcome to PowerPath Fundamentals.

Copyright 1996, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013 EMC Corporation. All Rights
Reserved. EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without
notice.
THE INFORMATION IN THIS PUBLICATION IS PROVIDED AS IS. EMC CORPORATION MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY
KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.
EMC2, EMC, Data Domain, RSA, EMC Centera, EMC ControlCenter, EMC LifeLine, EMC OnCourse, EMC Proven, EMC Snap, EMC SourceOne,
EMC Storage Administrator, Acartus, Access Logix, AdvantEdge, AlphaStor, ApplicationXtender, ArchiveXtender, Atmos, Authentica,
Authentic Problems, Automated Resource Manager, AutoStart, AutoSwap, AVALONidm, Avamar, Captiva, Catalog Solution, C-Clip, Celerra,
Celerra Replicator, Centera, CenterStage, CentraStar, ClaimPack, ClaimsEditor, CLARiiON, ClientPak, Codebook Correlation Technology,
Common Information Model, Configuration Intelligence, Configuresoft, Connectrix, CopyCross, CopyPoint, Dantz, DatabaseXtender, Direct
Matrix Architecture, DiskXtender, DiskXtender 2000, Document Sciences, Documentum, elnput, E-Lab, EmailXaminer, EmailXtender,
Enginuity, eRoom, Event Explorer, FarPoint, FirstPass, FLARE, FormWare, Geosynchrony, Global File Virtualization, Graphic Visualization,
Greenplum, HighRoad, HomeBase, InfoMover, Infoscape, Infra, InputAccel, InputAccel Express, Invista, Ionix, ISIS, Max Retriever, MediaStor,
MirrorView, Navisphere, NetWorker, nLayers, OnAlert, OpenScale, PixTools, Powerlink, PowerPath, PowerSnap, QuickScan, Rainfinity,
RepliCare, RepliStor, ResourcePak, Retrospect, RSA, the RSA logo, SafeLine, SAN Advisor, SAN Copy, SAN Manager, Smarts, SnapImage,
SnapSure, SnapView, SRDF, StorageScope, SupportMate, SymmAPI, SymmEnabler, Symmetrix, Symmetrix DMX, Symmetrix VMAX,
TimeFinder, UltraFlex, UltraPoint, UltraScale, Unisphere, VMAX, Vblock, Viewlets, Virtual Matrix, Virtual Matrix Architecture, Virtual
Provisioning, VisualSAN, VisualSRM, Voyence, VPLEX, VSAM-Assist, WebXtender, xPression, xPresso, YottaYotta, the EMC logo, and where
information lives, are registered trademarks or trademarks of EMC Corporation in the United States and other countries.

All other trademarks used herein are the property of their respective owners.

Copyright 2013 EMC Corporation. All rights reserved. Published in the USA.

Revision Date: April, 2013
Revision Number: MR-1WP-PWRPFD_2.0.

1 PowerPath Fundamentals
The objectives for this course are presented here. Please take a moment to review them.
PowerPath Fundamentals 2
The objectives for this module are shown here. Please take a moment to review them.
PowerPath has traditionally been viewed only as path management software supporting load
balancing and failover. With the addition of migration and encryption functionality, PowerPath has
expanded its scope of functionality as well as its supported platforms. Centralized licensing platforms
as well as standardized interfaces allow for the bundling of PowerPath Products together.
PowerPath should now be viewed as a family of technology having two distinct functions. Two
Products support the core function of Path Management and Optimization, while the other two
products perform a Data Protection function:

PowerPath Multipathing, sometimes referred to simply as PowerPath, is the same trusted product it
has always been. The product automates data path management and failover and recovery, as well as
optimizes load balancing to ensure application availability and performance. PowerPath Multipathing
runs on Windows, Solaris, HP-UX, and Linux.
PowerPath/VE automates and optimizes data path pools in virtual environments such as VMware to
ensure business process availability and performance and the ability to scale-out mission-critical
applications.
PowerPath Migration Enabler leverages the same underlying technology as PowerPath Multipathing
to enable data mobility technologies to perform data migrations to and from any network or
supported array type. This includes EMC and qualified non-EMC storage. The product allows
completion of these tasks without application downtime.
PowerPath Encryption, sometimes also called VNX Host Encryption, uses software agents on hosts and
centralized RSA Data Protection Manager appliances for key management to secure data from the
host to disk on storage.

Here is a list of the main business challenges that IT professionals face today.
Customers need their businesses up and running 24x7 in order to protect revenue streams,
support new product development efforts, and increase customer satisfaction.
Businesses need a method to manage their storage in order to maximize their technology
investments. IT professionals are searching for new and more efficient, policy-based,
automated management tools to manage and reduce the complexity of their environment.
Tools must be capable of tuning the environment in order to maximize server, storage, and
data path utilization.
These management tools must have the ability to quickly respond to various and changing
business requirements, provide continuous data protection, and tune the I/O streams of
concurrently running applications. These requirements are even more demanding as physical
hosts are redeployed as virtual hosts on enterprise virtual server environments such as
VMware vSphere and Microsoft Hyper-V.

Now that weve gone over the business challenges, lets see how PowerPath can help to solve
these problems. PowerPath can automatically tune host-to-storage connections when it
detects imbalances. It does this by selecting alternate paths for the data to be routed through.
It also enhances multi-path I/O capability in the SAN using automatic I/O load balancing and
online path failover functionality. These features are combined into one integrated package
known as PowerPath Multipathing, or simply PowerPath.
PowerPaths automatic load balancing functionality guarantees maximum use of multiple
paths to the data. Critical server and array resources, such as memory and processing power,
are not limited to using congested paths to the data while other paths to that same data are
under utilized. By maximizing the use of multiple data paths, server and array investments are
better allowed to be used to their full potential.
PowerPaths automatic path fail-over capability guarantees application availability. When the
host and storage environment are configured with redundant paths through multiple HBAs
and SAN fabrics, PowerPath automatically uses alternate paths to the device and continues to
balance I/O load across the remaining paths. When a path is restored, PowerPath
automatically adds the path back into the available paths list and puts it back into use for load
balancing.
When using PowerPath, all load balancing and failover policies are tunable without
interrupting application operations. Path management policies may be customized on the
host at the array level or on a device by device basis as required.

PowerPath can add value in many ways. For example, it is supported on many operating
system platforms including Windows, Linux, Solaris, HP-UX, and AIX and supports all EMC
arrays and several non-EMC arrays (referred to as third-party arrays).
PowerPath helps maximize SAN performance by using all the I/O processing and bus capacity
of all available paths to a device. Load balancing is administered on a host-by-host basis and
works by equalizing I/O load across all paths to the array from the host. PowerPath decisively
handles each I/O request by first checking current load balancing and path failover settings to
understand how to select the least-burdened available path.
PowerPath reduces management time and application downtime, because administrators no
longer need to maintain and configure paths statically across logical devices. With PowerPath,
no setup time is required, and paths are always configured for optimum performance.
If there is a failure in the data path, PowerPath fails over the I/O to an alternative path.
PowerPath determines the best and most efficient way to access devices and paths to them
across the SAN.
These are the key points covered in this module. Please take a moment to review them.
The objectives for this module are shown here. Please take a moment to read them.
PowerPath is host-based software that provides storage I/O path management. It operates
with various operating systems and storage arrays to intelligently manage the I/O paths which
connect them.
The illustration shows a typical PowerPath configuration where there are multiple paths
between the host and storage devices in the array. Each host path has its dedicated host bus
adapter (HBA) port, cable and Fibre Channel or iSCSI switch port. Through a switched SAN, it is
possible for any host HBA port to establish a connection to any storage device across a wide
selection of path components between the SAN switch and storage system interface port.
This is typically done with Fibre Channel Zoning or iSCSI Discovery Domains which limit the full
number of available pathways which PowerPath can take advantage of through the SAN fabric.
Due to operating system limitations, without path management software installed there can
be only one active path to each storage device.

When I/O path components are placed under management of path failover and load
balancing software such as PowerPath, the ability to access devices through multiple paths
increases both application availability and I/O efficiency. To best take advantage of this
benefit, the devices should be assigned equally across all available storage array interface
cards. This ensures that the full capacity of all array interfaces are being used.

PowerPath operates between the applications and the HBA driver . This position in the I/O
stack gives PowerPath the ability to intercept and manage all I/O between the applications
and their storage devices.
PowerPath intelligently manages the paths to a device by sending I/O down the optimal path
based on the current internal load balancing and failover policy setting for the device.
PowerPath also takes into account path usage and availability before deciding which path to
use for sending I/O. If a path to the device fails, PowerPath automatically reroutes the I/O
down an alternate path.
During installation, PowerPath determines all the paths to a device and builds a table as decisions are
made about how best to route I/O. PowerPath references this table and drives the I/O to either a
native or a pseudo device. The supported operating systems for these device types are listed on this
illustration.
The operating system creates native devices to represent and provide access to logical devices. A
native device is path-specific and represents a single path to a logical device. The device is native in
that it is provided by the operating system for use with applications. In this illustration, there is a
native device for each path. The storage system is configured with two shared logical devices, each of
which can be accessed by four paths. There are eight native devices, four in green (numbered as 0, 2,
4, and 6) that represent a unique path set to logical device 0; and four in blue (numbered as 1, 3, 5,
and 7) that represent a unique path set to logical device 1. When PowerPath is installed, it uses the
native device to point to all the paths to the device.
Applications do not need to be reconfigured to use native devices. The idea is to continue to use the
existing disk devices created by the operating system.
A PowerPath pseudo device represents a single logical device and the set of all paths leading to it,
which can contain any number of physical paths. Since a pseudo device represents all possible paths,
there is exactly one pseudo device per path. The example on this illustration shows that the logical
devices 0 and 1 are referred to by pseudo device names emcpower1c and emcpower2c. Each pseudo
device represents the set of paths connected to its respective logical device. For example,
emcpower1c represents the set of paths connected to logical device 0 and emcpower2c represents
the set of paths connected to logical device 1. In most cases, the application must be reconfigured to
use pseudo devices; otherwise, PowerPath load balancing and path failover functionality are not
available.
Active-active means all interfaces to a LUN are active simultaneously. In an active-active storage array,
if there are multiple interfaces to a logical device, they all provide equal access to the logical device. In
a configuration that includes an active-active array, PowerPath can spread the work load across all
paths simultaneously. In addition, PowerPath can failover across both paths to the LUN.
EMC Symmetrix, VMAX, IBM ESS, Hitachi TagmaStore, are examples of active-active arrays.
EMC CLARiiON and VNX block are examples of active-passive arrays. In the active-passive storage
system, SP A is designated as the primary or active route to the device, and therefore all I/O is directed
down the paths through SP A to the device. PowerPath load balances I/O across these active paths as
shown by the green arrows on this illustration.
The LUN can also be accessed through SP B but only after the device has been re-assigned (or
trespassed) to storage processor B. This path is referred to as a passive path, and PowerPath does not
send I/O down passive paths. Passive paths are shown by the blue arrows on this illustration.
If all active paths to a LUN become unavailable due to an SP error, alternate paths to that same LUN,
through the other SP, must be used. These will now be the active paths. This active path reassignment
is initiated by the interface of the other SP which will now be used to access the LUN. If PowerPath is
installed on the host accessing this LUN, the active path reassignment is initiated instead by
PowerPath, which instructs the storage system SP to make the reassignment. On a VNX or CLARiiON
array, these reassignments are known as trespassing. Trespassing can take several seconds to
complete. However, I/Os do not fail during this time. After devices are trespassed, PowerPath detects
the changes and seamlessly sends data via the new route. After a trespass, logical devices can be
trespassed back to their original paths once the error is corrected. If the VNX or CLARiiON array is
using ALUA, the trespass operation is avoided while switching LUN access via the non-owning SP.
PowerPath has several licenses available.
The Full PowerPath license permits the user to take advantage of the full set of PowerPath
load balancing and path failover functionality. PowerPath Enterprise and PowerPath
Enterprise Plus are other names for the fully-licensed product which are no longer used.
PowerPath/VE is the path management solution specifically for VMware ESX virtual server
environments. It is licensed according to both the number of physical VMware ESX hosts and
the number of CPU sockets per host. PowerPath/VE provides the full set of load balancing
and path failover functionality.
A PowerPath SE license supports back-end failover only. In other words, only paths from the
switch to storage arrays are candidates for failover. Back-end storage failover is for a single
HBA that has one path to the switch and is zoned to have up to two paths to the storage array.
Unlicensed PowerPath installations support EMC arrays only and has the same single-HBA
host limitation as PowerPath SE. When using unlicensed PowerPath, you may enter a license
at any time to enable PowerPath capabilities based on that license.
Evaluation, or trial licenses enable full PowerPath capabilities for a limited period of time.
You may come across other names for PowerPath, such as PowerPath Fabric Failover,
PowerPath Storage Processor Failover and PowerPath as included in the CLARiiON Utility Kit.
These variants are all equivalent in function to PowerPath SE.

This table illustrates the broad range of operating systems, clustering software, storage arrays,
logical volume managers, and filesystems that PowerPath supports. For current
interoperability information, always refer to the E-Lab Interoperability Navigator which is
available at the EMC Online Support website.
This table lists some of the most commonly used products which provide similar functionality
to PowerPath.
OS vendors provide native multipathing options in the form of multipath I/O management
offerings (MPIO) embedded in the OS. Because these are created by the OS vendor as distinct
frameworks or products, each OS has its own particular version of MPIO. OS Native MPIO
solutions may not offer such features as tunable performance, optimal data path utilization,
or automated load balancing and failover.
Refer to the Table here and the notes for detailed comparisons to each of these competing
products. One thing to note is that PowerPath 5.0 and later has added support for the
Microsoft DSM active/active load balancing capability. Leveraging this and some other native
MPIO enhancements allows PowerPath to continuously improve and work better with each
OS. PowerPath/VE, specifically for hypervisor hosts, adds the robust path management
features of standard PowerPath to the vitalization servers storage path environment.
Replacing the native path management with PowerPath greatly benefits all virtualized
application servers running on the hypervisor.
Here is a summary of PowerPath main features, which are discussed with specific examples
throughout this module.
Copyright 2013 EMC Corporation. Do not copy - All Rights Reserved. 20 PowerPath Fundamentals
The illustration shows that an application running on a server has four paths to the storage.
However, only one path is used because the volume manager native to the operating system
running on the server only allows one path to be defined. This is expressly done when the
administrator chooses a device name on which to build the volume group and file system.
Without PowerPath or another path failover software package, the loss of a channel means
one or more applications may stop functioning. This can be caused by the loss of an HBA,
storage array front-end port, Fibre Channel switch or a failed cable. In a standard non-
PowerPath environment, these are all single points of failure. In this case, all I/O that was
heading down the path is now lost, resulting in an application failure and the potential for
data loss or corruption.

Copyright 2013 EMC Corporation. Do not copy - All Rights Reserved. 21 PowerPath Fundamentals
It is shown here how PowerPath failover works on a host attached to an active-active array,
such as VMAX.
When a failure occurs, PowerPath transparently redirects the I/O down the most suitable
alternate path. The PowerPath filter driver looks at the set of paths to the device, considers
current workload, load balancing, and device priority settings, and chooses the best path to
send the I/O down.
The example on this illustration shows that PowerPath has three remaining paths to redirect
the failed I/O and to load balance across.
Because PowerPath spreads the additional load across the remaining channels, the system
continues to operate. Even in this degraded mode, the applications are still running, and just
as importantly, the pressure is off the administrative staff to make emergency corrections
immediately. Repairs can be postponed to coincide with a scheduled maintenance window or
other pre-determined time when host downtime will not impact the application users.

Path failover on an active-passive array is controlled by PowerPath in combination with array
technology. The illustration on this slide shows failover scenarios on an active-passive array. The logical
device is assigned to Storage Processor B (SP B). PowerPath is load balancing the application I/O across
the two ports on SP B. The paths to the device through SP A are considered passive and therefore not
supporting I/O. Let us consider what happens when a failure occurs on any of the components that
make up the path to the logical device.

Path Fault: If a backend path component becomes non-operational, PowerPath automatically fails the
path over to an alternate path. The logical device is not trespassed because an alternate path to the
logical device through the primary storage processor, SP B, is still available. Once the path is restored,
PowerPath automatically recognizes this and resumes sending I/O down the original path.

HBA Fault: If an HBA is lost, PowerPath fails the entire path over to an alternate HBA. In this case, a
trespass occurs because there is no path from the alternate HBA to the primary processor, SP B. The
trespass could be avoided if the HBA is zoned to SP B. When the HBA is brought back online,
PowerPath recognizes this and resumes sending I/O through the original HBA.

Storage Processor Fault: If SP B fails, PowerPath stops all I/O to SP B and trespasses the device over to
SP A. All I/O will be sent down the paths to SP A. When SP B is brought back online, PowerPath
recognizes that SP B is now available and resumes sending I/O to SP B.
PowerPath ensures that applications remain continuously available in the event of any of the above
scenarios.
PowerPath SE provides single HBA configuration support, back-end failover support, and non-
disruptive upgrade support.
The SE version has no multi-pathing or load balancing capabilities and therefore should not be
used in a high availability environment.
PowerPath Storage Processor Failover is the equivalent of PowerPath SE without a license key
that provides only basic failover functionality. PowerPath Storage Processor Failover protects
against VNX and CLARiiON SP failures. Also known as PowerPath Fabric Failover, it protects
against Symmetrix FA port failures, back-end storage system failures, and supports non-
disruptive upgrade (NDU) of storage system software.
If a failure occurs in an SP or an FA port, the response is to fail over (or transfer) the I/Os to a
different storage array port. The example on this illustration shows how PowerPath sends an
I/O down the path to a storage port: A failure occurs on the backend. PowerPath receives the
error and resends the I/O down the path to another storage port which has connection to the
required device.

Note that PowerPath SE is not a separate version of PowerPath software. In fact, all
PowerPath versions (with the exception of PowerPath/VE) are available within the single
instance of PowerPath when it is installed. PowerPath SE is effectively enabled just by
installation of an appropriate license key.

In this illustration there are several applications running on a server which have four paths to
the storage. However, only one path to a single storage device is used because the volume
manager native to the operating system only allows for one device path to be defined. This
one path is actually determined when the administrator chooses a device name on which to
build the volume group and filesystem.
The administrator configures the storage system to spread I/O load across the paths based on
snapshot measurements, guesstimates of average loading, predictive loads, and the
experience of storage experts. The final configuration has some applications sharing devices
down one path.
The example on this illustration depicts a snapshot of the system at a moment-in-time when
the depth of the I/O queues is very unbalanced. Host applications sitting on top of deep
queues are not achieving the data access efficiency they need.
In this instance, two of the applications are currently generating high I/O traffic, causing two
channels to become overloaded, as depicted by the red disks and the pending request stack.
The two other channels are lightly loaded. Eventually, the requests are handled and the
system returns to a more balanced load. In the meantime, the applications are data starved
and the users may experience less than optimal performance.
To solve the problem, the System Administrator has to manually reconfigure the system to
better balance the load, requiring application downtime. In any system, there are times when
the load is unbalanced due to an application experiencing heavy I/O requirements.

With PowerPath installed, applications will transparently access PowerPath instead of the HBA
driver directly. PowerPath distributes I/O requests across all available channels, reducing
bottlenecks and improving performance.
The diagram shows a situation similar to the previous slide, this time using PowerPath to
manage the multiple channels to minimize the I/O queue depth on all channels.

This allows PowerPath to constantly tune the server to adjust to changing loads from the
applications running on the server. This, in turn, improves the performance of the server by
enabling it to make better use of the storage. For every I/O, the PowerPath filter driver looks
at the volume path set and selects the path based on the load balancing policy and failover
setting for the device. The result of this is better application performance with fewer
operational resources spent on the care and maintenance of the system.
PowerPath does not manage the I/O queues themselves - that is still done by the HBA drivers
- but it directs the placement of I/O requests into those queues.
This illustration lists current PowerPath load balancing policies. Please take a moment to review them.
Symm_opt, CLAR_opt, or Adaptive policies choose paths based on path load and logical device
priority. Symm_opt (so) is the default for Symmetrix/VMAX arrays and CLAR_opt (co) is the default for
CLARiiON/VNX arrays. Adaptive (ad) is the default priority used with non-EMC arrays and performs
similarly to Symm_opt and CLAR_opt.
In Round Robin (rr) policy, I/O requests are assigned to each available path in rotation. It can be used
for all arrays that PowerPath supports.
Least I/Os (lo) policy load balance is based on the number of pending I/Os. I/O requests are routed to
the path with the fewest queued requests, regardless of total block volume. Least Blocks (lb) policy
load balancing is based on the number of blocks in pending I/Os. I/O requests are routed to the path
with the fewest queued blocks, regardless of the number of requests involved. Both policies can be set
on all arrays that PowerPath supports.
Request (re) uses the path that would have been used if PowerPath had not been installed. For pseudo
devices, it uses one arbitrary path for all I/O. For all devices, path failover is in effect. However, load
balancing is not in effect. No Redirect (nr) policy has neither load balancing nor failover in effect. This
policy is valid for all arrays that PowerPath supports. Both policies can be used on all arrays.
Finally, basic failover policy (bf) has no load balancing in effect. I/O routing on failure is limited to one
host bus and one port on each Storage Processor. This policy is required for a non-disruptive upgrade.
It protects against Storage Processor and back-end failures, but not against HBA or host loop failures.
Basic failover is valid for Symmetrix, CLARiiON, VNX and HP EVA arrays.
Please consult the latest PowerPath Product Guide and Release Notes for the most current
information about load balancing polices and their use.

The autoprobe function periodically probes inactive paths to identify failed paths before
sending host I/O requests. This process allows PowerPath to proactively close paths before an
application experiences a time-out when sending I/O over failed paths.
The autoprobe function uses SCSI Inquiry commands for the probe so that even a not-ready
device returns successfully. It tests one LUN per HBA and port connection roughly every 30
seconds and skips LUNs that have received I/O in the last 30 seconds. If one probe fails down
a path, it kicks off error checking routines. For failed paths, these routines check every LUN
down that path to proactively close paths.
Autorestore runs every five minutes and probes every failed or closed path to determine if it
has been repaired. Like autoprobe, it uses SCSI Inquiry commands.
Using a combination of autoprobe and autorestore features means that if a path fails, it is not
used but the application continues to run. Autorestore then tests the paths viability and
automatically restores the connection in approximately five minutes. Administrators can also
restore paths manually using the PowerPath CLI or Administrator GUI.
Priority settings are seldom used and have no effect unless the policy in use is Symmetrix
Optimize or CLARiiON Optimize.
The illustration shows the result of setting a device priority of 10 for LUN K.
Two applications are running on the host Application A and Application B. Application A has
data on LUN K. Application B is using one of the other LUNs as shown on this illustration.
Before setting the device priority, both applications had equal access to all paths to the LUNs.
After setting the device priority on device K to 10, Application A is given a higher priority to
access the LUN so that more of Application As I/Os are queued up on that path. Application
Bs I/Os are forced to use the remaining paths to the devices.
The device priority is an integer in the range of 0 to 10, with the default being 0. The higher
the number, the higher the priority. Setting all devices to the same priority nullifies the effect
of the command.
Setting device priorities is useful when certain applications require highly responsive and
uninterrupted access to a device. For example, a database log disk that requires I/Os to be
satisfied quickly, otherwise application performance suffers.
Note that as of the current release, the ability to set the I/O priority for a device is currently
not a recommended configuration, refer to the EMC PowerPath Family CLI and System
Messages Reference document.
The illustration shows how PowerPath channel groups can be used to tune I/O performance.
PowerPath provides the flexibility to define paths to a device as active or standby. A standby
path is only used when all active paths to a device fail. PowerPath considers active paths as
available for load balancing and failover. Paths can be dynamically added and removed by
setting them to standby or active mode. For example, you have eight paths configured as four
dedicated to one application and four dedicated to another. When one application needs
more bandwidth (for example, the duration of nightly batch processing), you can
automatically add more paths through scripts triggered during the batch processing time
window. These paths can later be assigned back to where they were when the batch
processing jobs are finished. Dedicated paths to specific applications are called Channel
Groups.
In this example, there are two applications running on the host Application A and
Application B. The server is configured for six paths to the devices. Both applications are very
busy. You decide that you do not want the activity of one application to impact the
performance of the other. You set two paths to the devices used by Application A into standby
mode. These paths are shown in yellow. Application A now has four active paths to the
devices. To further isolate the I/O for both applications, you use PowerPath to set four of the
paths to the devices used by Application B to standby mode. These paths are represented in
green. Now Application A has four active paths and Application B has two active paths to the
devices.
Open Systems clustering technology manages application availability by detecting failures and
restarting high-availability applications on a surviving cluster node.
The deployment of PowerPath in the cluster eliminates the application downtime due to a
path or channel failure. PowerPath detects the channel failure and uses alternate channels so
that the cluster software does not have to reconfigure the cluster to keep the applications
running.
PowerPath improves the availability of the applications running in the cluster.
Many clusters are deployed to provide performance scalability and intelligent load balancing
can help the administrator maximize performance to realize the greatest value from their high
availability clustering investment.
PowerPath must work in cooperation with the cluster software. This does not require special
code in the cluster, but does require PowerPath to work with the storage management
component of the cluster software. Validating PowerPath with the cluster software systems
requires extensive knowledge and testing of the cluster software with PowerPath. For current
interoperability information, always refer to the E-Lab Interoperability Navigator which is
available at the EMC Online Support website.
The PowerPath Administrator graphical user interface (GUI) for Windows consists of two
applications the PowerPath Monitor and the PowerPath Administrator GUI.
The PowerPath Monitor continuously monitors the PowerPath configuration and alerts you
for status changes.
The PowerPath Administrator GUI is used to configure, monitor, and manage PowerPath
devices.
This illustration shows the PowerPath Administrator Console. The PowerPath Administrator
has two panes. On the left is the Scope pane, where PowerPath objects are displayed in a
hierarchical list that can be expanded or collapsed. On the right is the Results pane that
provides a view of configuration statistics for PowerPath objects selected in the Scope pane.
The CLI interface is available for all supported OS. The commands are used to view or change
the PowerPath options for the HBA port and the path of the device.
This console screenshot shows the output of the powermt help command. This command will
display all available parameters and qualifiers which may be used with the powermt
command. Please take a moment to review this information.
PowerPath Viewer is a free utility that provides centralized, remote monitoring of your
PowerPath-managed storage environment. PowerPath Viewer comprises two main
components: It provides a consolidated display of events and allows you to view and monitor
PowerPath-managed hosts through a graphical user interface (GUI). You can view hosts, host
groups, LUNs, individual paths to each LUN, and buses.
All host discovery that takes place within PowerPath Viewer Console is for viewing and
monitoring within PowerPath Viewer only. Any host group that you may create within
PowerPath Viewer Console does not affect the storage configuration within the hosts
themselves. So you cannot manage/administer your PowerPath Hosts from the PowerPath
Viewer, only monitor those hosts.
The information displayed in PowerPath Viewer Console is similar to the information that is
displayed when you run the powermt display PowerPath command. It is also similar to the
information found in the PowerPath GUI console on Windows. PowerPath Viewer Console
presents the information in windows and panes, called Views.

PowerPath Viewer provides a single solution for easily monitoring configurable alerts for up
to hundreds of PowerPath Hosts. This utility uses the type of data normally returned with the
powermt display command from each monitored PowerPath host and displays the
consolidated information on a centralized management station. Users can see the big picture
of their storage infrastructure, or drill down for details. Information is displayed in a user-
friendly GUI.
From a single console, users can easily keep track of path, bus and storage device events for
each PowerPath host. Any changes in the path state can be included in an email so that an
administrator is notified.
This tool provides clear business benefits as less time and fewer resources are required to
monitor a complex storage infrastructure.

There are two components required for PowerPath Viewer to operate.
The first component is the PowerPath Management Component that runs on each PowerPath host
monitored by the PowerPath Viewer. The PowerPath Management Component is the means through
which the remote hosts and PowerPath Viewer Console communicate.
The Management Component is bundled with PowerPath on the following platforms: Windows, Linux,
Solaris, and HP-UX. For AIX, the Management Component is located on Powerlink as a separate
download. The Management Component is not required for PowerPath/VE for VMware because the
PowerPath Viewer Console communicates with the remote vSphere host via the Message Passing API.
You must discover the PowerPath/VE hosts through a similar discovery process as for normal
PowerPath hosts. Instead of a heartbeat, the Viewer polls the VMware vSphere host to check for
device status change.
The Management Component listens for kernel events from the PowerPath driver, sends alerts to the
Console over IP, and listens for commands from the Console, among other tasks.
The second required component is the Java-based PowerPath Viewer Console that consolidates all of
the PowerPath hosts information and displays it in a user-friendly GUI. The GUI is ECUE compliant and
has the same look and feel as other EMC management software tools.
The Viewer Console and remotely managed PowerPath hosts communicate over IP. Neither a Fibre
Channel, iSCSI, nor an FCoE connection is required for communication. PowerPath Viewer Console
also alerts you to any changes in status to PowerPath devices through two monitors: the Path Alert
Monitor, for changes in the status of physical data paths between hosts and individual LUNs; and the
Bus Alert Monitor, for changes in the status of SAN connections between hosts and arrays.

The Viewer Console is supported on Windows 7 or Windows Server 2008R2, physical or
virtual. The supported architectures are X86 or X64.
In the current version, the PowerPath Viewer can monitor up to 500 hosts. Each newly
installed PowerPath Viewer Console must be configured to discover its own set of PowerPath
hosts.

The current console architecture allows a single PowerPath host to be monitored by a
maximum of five PowerPath Viewer Consoles.

Minimum system requirements for the PowerPath Viewer Console to monitor are listed here.
PowerPath Viewer is currently supported for Windows hosts with PowerPath v5.7 and later.
EMC does not recommend running the PowerPath Viewer Console on any host running
PowerPath.
Any arrays qualified to work with PowerPath v5.3 and later are qualified for PowerPath
Viewer. See the E-Lab Interoperability Navigator, available on Powerlink, for up-to-date
information.

These are the key points covered in this course. Please take a moment to review them.
This concludes the training.

Powerpath Fundamentals - SRG PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Powerpath Fundamentals - SRG PDF

Încărcat de

Drepturi de autor:

Formate disponibile

Copyright 2013 EMC Corporation. Do not copy - All Rights Reserved.

Welcome to PowerPath Fundamentals.

S-ar putea să vă placă și