VTrak E-Class and Falconstor VTL Best Practices Guide

Best Practices Guide
Promise VTrak E610f External Disk Array Subsystem and
FalconStor VTL
November 2, 2009
Table of Contents
About This Guide .................................................................................................................................................................. 2
Introduction ........................................................................................................................................................................... 2
Guidelines .............................................................................................................................................................................. 3
Choosing a Suitable Topology ............................................................................................................................................ 4
DAS..................................................................................................................................................................................... 4
HA Options for DAS........................................................................................................................................................ 4
Storage Level ............................................................................................................................................................. 4
Appliance Level.......................................................................................................................................................... 4
SAN..................................................................................................................................................................................... 5
HA Options for SAN........................................................................................................................................................ 5
Access Control in a SAN Topology ................................................................................................................................ 5
Configuring the VTrak .......................................................................................................................................................... 6
Single/Dual Controller Module ............................................................................................................................................ 6
Memory Option ................................................................................................................................................................... 6
Hard Drive........................................................................................................................................................................... 6
Recommended Settings ..................................................................................................................................................... 6
Alert Notification.................................................................................................................................................................. 7
Configuring the RAID............................................................................................................................................................ 8
RAID 5 and RAID 6............................................................................................................................................................. 8
Global Revertible Hot Spare ............................................................................................................................................... 9
Logical Drive Attributes....................................................................................................................................................... 9
Initializing the Logical Drives ............................................................................................................................................ 10
Setting up Access Control ................................................................................................................................................. 10
Assigning Storage to VTL ................................................................................................................................................. 10
Enabling LUN Mapping................................................................................................................................................. 10
Configuring LUN Mapping ............................................................................................................................................ 11
Assign VTL Tape to VTL Clients....................................................................................................................................... 11
Tuning the Performance..................................................................................................................................................... 11
Distributing I/O Load among Subsystems ........................................................................................................................ 11
Distributing I/O Load on Logical Drives ............................................................................................................................ 11
Distributing I/O Load on Fibre Channel Ports................................................................................................................... 11
Identifying and Removing Slow Drives ............................................................................................................................. 11
Summary .............................................................................................................................................................................. 12
About Promise..................................................................................................................................................................... 13
Contact Us ........................................................................................................................................................................... 13
List of Figures
Figure 1. DAS topology ........................................................................................................................................................... 4
Figure 2. SAN topology ........................................................................................................................................................... 5
Figure 3. Recommended RAID configuration ......................................................................................................................... 8
Figure 4. How global revertible hot spares work..................................................................................................................... 9
List of Tables
Table 1. Recommended settings ............................................................................................................................................ 7
Table 2. Recommended settings for logical drive attributes ................................................................................................. 10
Best Practices Guide
1
About This Guide
This Best Practices Guide describes the recommended VTrak E610f configurations and tips for achieving the best
performance with the VTrak E610f and the VTL appliance. It is intended for system administrators, storage architects,
system integrators, and resellers who are planning to install or qualify FalconStor VTL using Promise VTrak E610f.
Introduction
The explosive growth of data generated by businesses continues to bring challenges in the area of backup and recovery
operations. Increasing volumes of data and narrowing backup windows have made traditional tape backup increasingly
difficult to maintain. In addition, tape backup has often led to longer recovery periods, leaving more room for media and
mechanical failures. Businesses require faster and more reliable ways to back up and recover their data.
Virtual tape appliances, such as FalconStor Virtual Tape Library (VTL) Storage Appliances and Enterprise Edition,
address the challenges associated with tape backup. FalconStor VTL solutions offer fast backup, data deduplication,
enterprise-wide replication, and tape integration in one solution, without requiring changes to the existing environment.
Built-in data deduplication features automatically and transparently eliminate redundant data, extending data retention and
allowing users to reduce or even eliminate investments in physical tape media.
However, FalconStor VTL solutions require massive amounts of storage that must scale both in performance and in
capacity. In the market, many existing disk array products are designed for traditional enterprise transactional computing.
They cannot provide the scalability that VTL applications require, and they are usually too expensive to deploy in large
volume.
Promise VTrak E610f external disk array subsystem was designed to work seamlessly with VTL applications, providing
exceptional input/output (I/O) performance for concurrent VTL operations like backup, restore, and deduplication with a
host of features, including:

Aggregated 1.6Gb/s bandwidth and low
overhead implementation

Dual 4Gb Fibre Channel host interface ports

Support for RAID levels 0, 1, 10, 1E, 5, 6, 50,
and 60

Robust error recovery

No single point of failure design

Active-active dual controllers
The VTrak E610f has 16 drive bays, and can support up to 80 SAS or SATA drives through Promise VTrak J610S
expansion units, or up to 80 terabytes (TB) of raw capacity (based on using a 1TB SATA drive). It offers a highly scalable,
reliable, and cost-effective storage solution for FalconStor VTL Appliances and Enterprise Edition, appropriate for small
and medium-size businesses (SMBs) to Enterprise customers.
The recommendations in this best practices guide were tested on the following hardware and software versions:

Promise VTrak E610f external disk array subsystem, firmware version 3.33.0000.00 (SR 2.5)

Promise VTrak J610s external disk subsystem, firmware version 1.07.0000.12

FalconStor VTL Server v5.1
Best Practices Guide
2
Guidelines
This section provides an overview of the topics discussed in this best practices guide for configuring and tuning the VTrak
E610f and FalconStor VTL to achieve optimal performance. Detailed instructions are provided in the corresponding
sections of this Guide.
Choosing a Suitable Topology and HA Option


DAS
SAN
Configuring the VTrak





Single/Dual Controller Module
Memory Option
Hard Drive
Recommended Settings
Alert Notification
Configuring the RAID




RAID 5 and RAID 6
Global Revertible Hot Spare
Logical Drive Attributes
Initializing the Logical Drives
Setting up Access Control


Assigning Storage to VTL
Assigning VTL Tape to VTL Clients
Tuning the Performance




Distributing I/O Load among Subsystems
Distributing I/O Load on Logical Drives
Distributing I/O Load on Fibre Channel Ports
Identifying and Removing Slow Drives
Best Practices Guide
3
Choosing a Suitable Topology
To support different VTL market segment requirements, the VTrak E610f offers two topology options for VTL: Direct
Attached Storage (DAS) and Storage Area Network (SAN). A DAS topology (shown in Figure 1) is a lower cost approach,
suitable for small to medium scale deployments. A SAN topology is massively scalable in terms of both capacity into the
petabyte range and of performance up to multiple terabytes per hour, for large-scale backup applications with stringent
backup window requirements. With both the DAS and SAN topologies, there are High Availability (HA) options that can be
configured at the storage (VTrak) and appliance (VTL) levels for meeting different application requirements.
DAS
Figure 1. DAS topology
VTL Appliance A
VTL Appliance B
(Option for HA)
In a DAS topology, the VTL appliances are
directly connected to the VTrak E610f through
their Fibre Channel ports.
VTrak E610f
The VTrak E610f can be connected to up to four Promise VTrak J610s expansion units.
VTrak J610s
Expansion Unit(s)
HA Options for DAS
There are two levels of high availability options for meeting different application requirements.
Storage Level
The VTrak E610f supports dual controller modules, which offer fully redundant paths to the VTL appliance with no single
point of failure. As shown in Figure 1 above, each VTL appliance has two Fibre Channel ports, each of which can be
connected to one Fibre Channel port on each of the storage controllers.
Appliance Level
This HA option offers redundancy at the VTL appliance level for the VTL cluster. If one of the VTL appliances fails, the
other VTL appliance takes over the VTL requests from the backup clients without service interruption. In this case, an
optional VTL appliance could be installed with two Fibre Channel ports, each connected to the remaining Fibre Channel
port on each of the storage controllers.
Best Practices Guide
4
SAN
Figure 2. SAN topology
VTL Appliance A
VTL Appliance B
VTL Appliance C
VTL Appliance D
FC Fabric
VTrak E610f
VTrak E610f
VTrak J610s
Expansion Unit(s)
VTrak J610s
Expansion Unit(s)
In a SAN topology (Figure 2), multiple VTL appliances are connected to multiple VTrak E610f units through one or more
Fibre Channel switches (collectively called an FC fabric). Similar to the DAS topology, the capacity of each VTrak E610f
could be expanded using VTrak J610s expansion units through SAS connections.
HA Options for SAN
A SAN topology also provides two levels of high availability options, similar to the DAS topology – at the storage level and
at the VTL appliance level. The difference is the appliances and storage are always connected through the Fibre Channel
switches instead of being directly connected. As illustrated in Figure 2 above, this topology offers completely redundant
paths from all VTL appliances to all storage subsystems, with no single point of failure.
Access Control in a SAN Topology
As the SAN topology places multiple VTL appliances and VTrak E610f units inside the same storage network, it is
important to configure the access control among them properly. This can be either achieved through zoning on the Fibre
Channel switch or through logical unit number (LUN) mapping on the VTrak E610f units. Refer to the section in this best
practices guide entitled “Setting up Access Control” for more details about how to configure the access control between
the VTL appliances and VTrak units.
Best Practices Guide
5
Configuring the VTrak
This section lists best practices for configuring your VTrak E610f with FalconStor VTL for optimal performance. In addition
to the best practices specified in this section, we recommend you also follow the general best practices for Promise VTrak
E-Class RAID solutions in the following knowledge base article.
10093 – Best practices for E-Class VTrak RAID
http://kb.promise.com/KnowledgebaseArticle10093.aspx?Keywords=best+practice
Single/Dual Controller Module
Depending on your application availability and performance requirements, you may choose either the single or dual
controller module configuration. To achieve high availability and optimal performance, Promise recommends using the
dual controller module configuration.
Memory Option
For high performance VTL applications, Promise recommends using two gigabytes of memory, on the controller,
and four-cell battery for ample cache memory backup retention.
Two gigabytes of memory and four-cell battery are not the default configuration of VTrak E610f.
When ordering, ask for the following SKUs:

Two gigabyte memory, SKU VTEMEM2G

Four-cell battery, SKU VTBAT4CELL
To place your order, please contact Promise Sales. See “Contact Us” on page 5.
Hard Drive
VTL is generally a high capacity application. For this reason, Promise recommends using high capacity, low cost
enterprise-capable SATA drives. Please refer to the Promise VTrak E610f compatibility matrix on the Promise website,
under Support > Compatibility Matrix, for compatible hard drive models.
Recommended Settings
The following table lists the recommended settings for the VTrak E610f for VTL and Single Instance Repository (SIR)
applications:
Best Practices Guide
6
Table 1. Recommended settings
Setting
Value
Rationale
Active-Active
Supports high availability and load balancing with VTL/SIR.
The default setting is Active-Active.
Disabled
Ensures compatibility with VTL/SIR.
The default setting is Disabled.
Subsystem
Redundancy Type
Controller
LUN Affinity
Force Read Ahead Enabled
Achieves the optimal performance for multiple stream
backup/restore and deduplication.
The default setting is Disabled.
Please refer to VTrak E610f Product Manual about how to
enable it.
Adaptive Write
Back Cache
Enabled
Protects the data integrity in case of Battery Backup Unit (BBU)
failure or a low battery condition.
The default setting is Enabled.
Auto Rebuild
Enabled
Enable Automatic Rebuild to replace bad drive or Predictive
Drive Migration (PDM) to replace a drive that is going to fail.
These actions improve data reliability by reducing or eliminating
periods of lost redundancy.
The default setting is Enabled.
SATA Physical Drive Settings
ReadAheadCache
Enabled
Helps to achieve the optimal performance for multiple stream
backup/restore and deduplication.
The default setting is Enabled.
Write Cache
Enabled
Helps to achieve the optimal performance for multiple stream
backup/restore and deduplication.
The default setting is Enabled.
Medium Error
Threshold
64
This value is the number of bad blocks tolerated before the
controller marks a physical drive as Dead. By enabling this
setting, you can identify problem drives before they fail, thereby
improving system performance and reliability.
The default setting is 0, meaning disabled.
Please refer to E610f Product Manual about how to set Medium
Error Threshold.
Also see the best practice at:
http://www.promise.com/support/support_eng1.asp,
And search for keyword “Medium Error Threshold.”
Promise recommends leaving all other settings at their factory default values.
Alert Notification
It is important to set up the alert notification on the Promise VTrak E610f external disk array subsystem, as this feature
notifies the system administrator and allows him to act accordingly in the event of any status change or failure condition.
The administrator should set up event notification either through Email or Simple Network Management Protocol
(SNMP) during the initial setup phase.
Best Practices Guide
7
Usually, Email is a preferred notification method. In certain user environments, there may be an existing alert notification
infrastructure that is based on a standard network management protocol such as SNMP. In this case, SNMP notification
can be configured to integrate with the existing infrastructure.
For detailed instructions on how to configure event notification through Email and SNMP, please refer to the VTrak E610f
Product Manual. You can find all of the Product Manuals on the Promise website under the Downloads section (Support >
Downloads).
Configuring the RAID
Promise VTrak E-Class external disk array subsystems support RAID levels of 0, 1, 1E, 10, 5, 6, 50 and 60. They also
support the mixing of different RAID level volumes in the same disk group to meet the requirements of different application
performance profiles.
This section provides the proven configurations for typical VTL applications to achieve the best backup and deduplication
performance.
Figure 3. Recommended RAID configuration
As illustrated Figure 3, in each enclosure, PD (physical drive) 1 through 15 form a disk array, and PD 16 serves as the hot
spare. In each disk array, multiple logical drives (volumes) can be carved out. Depending on the capacity requirement,
this configuration can be repeated on up to four VTrak J610s expansion units per VTrak E610f RAID head.
Depending on the application requirement, you may configure more drives as hot spares, or you may also configure fewer
hot spares.
RAID 5 and RAID 6
The recommend RAID configuration for the logical drives is RAID 6, as RAID 6 offers excellent reliability due to double
spindle failure protection. In the case that logical drives are used to store backup data temporarily for later deduplication
processing, the logical drives can be configured as RAID 5.
In either case, you should configure at least one hot spare drive for every 16 physical drives to protect the RAID
redundancy in case of drive failure.
Best Practices Guide
8
Global Revertible Hot Spare
To achieve the best reliability and manageability, Promise recommends configuring all of the spares as global
revertible hot spares. The advantage of global revertible hot spares is that each of the hot spares can be used for
rebuilding the same logical drive or different logical drives, in the same enclosure or in different enclosures. Therefore,
when there are unevenly distributed drive issues, there are still enough hot spares to provide coverage and to allow RAID
redundancy to be restored in the shortest time frame.
As revertible spares, when the rebuild completes and the failed drive is replaced with a new drive, the used spare will
automatically be freed up and become a revertible spare again after the reversion process, called Transition, initiates and
completes in the background.
Due to reliability considerations, Promise does not recommend mixing drives from different enclosures in the same
disk array. Therefore, when a rebuild completes through using a hot spare from a different enclosure, Promise
recommends replacing the failed drive as soon as possible. When the new drive is inserted, Transition will start
automatically in the background. Eventually, the disk array will consist of drives in the same enclosure again.
The follow diagram illustrates how global revertible hot spares work.
Figure 4. How global revertible hot spares work
Stage 1
Enclosure 1 - E610f
Assuming using RAID 6 configuration.
Everything started out optimal condition.
Enclosure 2 PD1 and PD2 were
replaced with new drives.
Transition instantly kicked in
automatically.
When Transition completed, the new
drives became part of disk array 2, and
Enclosure 1 and Enclosure 2 PD16 were
reverted back to hot spare.
The configuration recovered exactly
the same as the original stage
Enclosure 2 - J610s
Enclosure 2, PD1 and PD2 failed.
Stage 3
Stage 2
Enclosure 1 - E610f
Enclosure 2 - J610s
Enclosure 1 - E610f
Rebuild instantly kicked in
automatically using Enclosure
2 PD16 and Enclosure 1 PD16.
After rebuild completed, the
disk array 2 became crossing
enclosure 1 and 2.
Enclosure 2 - J610s
As illustrated in Figure 4, global revertible hot spares ensure overall system reliability in case of drive failure. In addition,
they also facilitate management by enabling the entire rebuild and Transition process to be automated. Once the process
completes, the system is restored to its original state.
Logical Drive Attributes
To achieve the best performance, Promise recommends configuring each of the logical drives with the attribute values as
listed in the table below:
Best Practices Guide
9
Table 2. Recommended settings for logical drive attributes
Attribute Name Value
Rationale
Stripe Size
64KB
Backup, restore, and deduplication I/O requests are usually
128KB, 256KB, or larger. A stripe size of 64KB can ensure the
VTL or SIR I/O request will stripe across as many drive
spindles as possible, hence achieving the best I/O
performance.
Read Cache Policy
Read Ahead
VTL restore or SIR read requests are sequential for single
streams. The Read Ahead setting boosts performance for
sequential I/O. Even in the case of multiple stream reads,
Promise lab tests indicated it boosted the performance
significantly, combined with Force Read Ahead enabled at the
controller level.
Write Cache Policy
Write Back
VTL backup requests are sequential write operations for
single stream. However, a VTL appliance can usually serve
multiple backup clients simultaneously. With multiple stream
VTL backup, the write access becomes random. By enabling
the Write Back cache, it avoids performance degradation due
to random drive access, therefore improving the performance
dramatically.
Other Attributes
Leave as default
Initializing the Logical Drives
After creating the logical drives, make sure to initialize all of them. This action ensures optimal performance.
Depending on the capacity of the logical drives, it may take 40 minutes to a few hours to initialize all of the logical drives.
Therefore, you are wise to plan ahead before running a performance test or placing a solution into production.
For detailed instructions about how to initialize logical drives, please refer to VTrak E610f Product Manual. You can find all
of the Product Manuals on the Promise website under the Downloads section (Support > Downloads).
Setting up Access Control
This section offers instructions and recommendations for how to assign storage to VTL as well as how to assign VTL tape
to VTL clients in a SAN topology. These recommendations apply to SAN topologies only.
Assigning Storage to VTL
Assigning storage to VTL can be achieved either through configuring zoning on the Fibre Channel switch or through
configuring LUN Mapping on the VTrak E610f. This best practices guide provides instructions on configuring LUN
Mapping on the VTrak E610f. For instructions on configuring zoning on the Fibre Channel switch, please refer to the Fibre
Channel switch User Manual.
Enabling LUN Mapping
By default, LUN Mapping is disabled on the VTrak E610f. When LUN Mapping is disabled, all of the logical drives on the
external array storage subsystem are accessible to all of the Fibre Channel initiator hosts. For a DAS topology or for a
SAN topology using zoning access control, LUN Mapping should be disabled.
However, for a SAN topology with access control set through LUN Mapping, LUN Mapping must first be enabled on the
subsystem before access control can be configured. Please refer to the VTrak E610f Product Manual for detailed
instructions on how to enable LUN Mapping.
Best Practices Guide
10
Configuring LUN Mapping
Once LUN Mapping is enabled, it can be configured per logical drive and per initiator, in other words, on a per VTL
appliance basis. To assign a logical drive to a VTL appliance, create a LUN map between the logical drive and the Fibre
Channel initiator ports on the VTL appliance.
For detailed instructions on how to configure LUN Mapping, please refer to the VTrak E610f Product Manual.
Assign VTL Tape to VTL Clients
A SAN topology is typically used to connect VTL appliances and VTL clients, where backup software has been installed.
Access control is required to configure which client has access to which VTL appliance and, more specifically, to which
VTL tape on the appliance.
Similar to assigning storage to VTL, the FalconStor VTL solution also offers a way to assign VTL tape devices to VTL
clients, similar to LUN Mapping on the VTrak E610f. For detailed instructions on how to assign VTL Tape to VTL clients,
please refer to the FalconStor VTL User Manual.
Tuning the Performance
This section lists tips for fine-tuning the VTrak E610f for optimal performance with the FalconStor VTL solution.
Distributing I/O Load among Subsystems
You can determine the I/O load on each subsystem through the real-time performance monitoring function in Promise
WebPAM PROe software. Promise WebPAM PROe software is a web-based graphical user interface embedded inside
each VTrak E610f. To access the graphical user interface, launch your Internet browser and type in the IP address of the
Management port; you will be connected to the Promise WebPAM PROe software login screen.
Through the performance monitoring function, you can see the I/O load of each subsystem and can determine whether
the loads are evenly distributed. If the loads are not evenly distributed, you can adjust the access control between the VTL
clients and the access control between the VTL and the VTrak E610f to balance the I/O load.
Please refer to the section in this best practices guide on Setting up Access Control for more details about how to
configure access control.
Distributing I/O Load on Logical Drives
Through the WebPAM PROe performance monitoring function, you can also find out whether the I/O load is evenly
distributed among the logical drives. If the load is not evenly distributed, you may adjust the access control between the
VTL and the logical drives of the VTrak E610f to balance the I/O load.
Please refer to the section in this best practices guide on Setting up Access Control for more details about how to
configure access control.
Distributing I/O Load on Fibre Channel Ports
You can use the performance monitoring function to find out whether the I/O load on the Fibre Channel ports connected
to the VTL is evenly distributed. If the load is not distributed evenly, you may change the preferred path to the logical
drives through the FalconStor VTL management console.
Please refer to the FalconStor VTL User Manual for detailed instructions on how to change the preferred path to the
logical drives from the storage subsystem.
Identifying and Removing Slow Drives
Not all SATA drives are equal. Some drives perform significantly slower than others, in term of response time and latency.
Any slow drive brings down the overall VTL solution performance. Therefore, Promise recommends stress testing the
drives during a pre-production test, then identifying and removing the slow drives before you deploy the solution
to the production environment.
You may also find out whether there are any slow drives using the WebPAM PROe performance monitoring function. The
WebPAM PROe real-time performance monitoring function is available in VTrak E610f firmware version 3.33.0000.00 and
Best Practices Guide
11
later. Please refer to the VTrak E610f Product Manual for more detailed instructions on how to use the WebPAM PROe
software and the real-time performance monitoring function.
Summary
The FalconStor VTL solution addresses the business challenge of making backup and recovery fast and reliable. Storage
plays a significant role in the reliability and performance of the VTL solution. Through efficient use of the Promise VTrak
E610f external disk array subsystem, the VTL backup and recovery performance becomes exceptional, the backup and
recovery windows are significantly shortened, and the total cost of the VTL solution is dramatically reduced.
Promise VTrak E610f external disk array subsystem is a versatile storage solution that can support many different
application requirements. However, for different applications, best practices may vary. In order to achieve the best
performance when planning FalconStor VTL installation with Promise VTrak E610f external disk array subsystem, please
follow the guidelines in this Best Practices Guide and refer to the additional resources specified in this paper.
Best Practices Guide
12
About Promise
With a long history of innovation, Promise Technology develops and manufactures sophisticated RAID solutions
recognized worldwide, ranging from a complete line of RAID controller cards to SAS/SATA RAID subsystems. Meeting
enterprise, mid-range, and entry-level data protection needs, Promise products are distinguished by their common RAID
code, unified management interface, and unparalleled support. Known as the originator of SATA/ATA RAID products,
Promise's comprehensive product base includes High Availability (HA) standalone RAID subsystems with standardsbased management interfaces, host-based (internal) RAID controllers, NAS appliances for SOHO, and SATA ASICs
integrated into an extensive list of industry-leading motherboards. Headquartered in Milpitas, California, Promise has
operations throughout Asia and Europe.
Contact Us
:: Promise USA
Milpitas, CA
1-408-228-1400
sales@promise.com
:: Promise EMEA
El Son, The Netherlands
31-40-235-2600
sales@eu.promise.com
:: Promise Taiwan
Hsinchu, Taiwan
886-3-578-2395
sales@tw.promise.com
:: Promise Germany
Dortmund, Germany
49-02-31-56-76-48-0
sales-de@eu.promise.com
:: Promise China
Beijing, China
86-10-8857-8085 or 8095
sales@cn.promise.com
:: Promise Italy
Rome, Italy
39-06-3671-2626
sales-it@eu.promise.com
:: Promise Japan
Tokyo, Japan
81-03-5333-3631
sales@jp.promise.com
:: Promise UK
Wokingham, England
44-0-870-112-59-77
sales@eu.promise.com
:: Promise Australia
Sydney, Australia
61-2-8065-2055
sales@au.promise.com
To learn more, visit www.promise.com
Best Practices Guide
13
© 2009 Promise Technology, Inc. The information contained herein is subject to change without notice. The only warranties for Promise products and
services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting
an additional warranty. Promise shall not be liable for technical or editorial errors or omissions contained herein.
Promise is a trademark of Promise Technology, Inc. All other brand and product names are registered marks of their respective proprietors.
Issue v1.1, Nov 2009
Best Practices Guide
14