Introduction To The EMC VNXe3200
Introduction To The EMC VNXe3200
Abstract
This white paper introduces the architecture and functionality of
the EMC VNXe3200. This paper also discusses some of the
advanced features of this storage system.
May 2014
Table of Contents
Executive Summary ................................................................................................. 5
Audience ............................................................................................................................ 6
Terminology ....................................................................................................................... 6
Hardware Overview................................................................................................ 10
Quad-Core CPU ................................................................................................................. 10
Disk Processor Enclosure (DPE) ........................................................................................ 10
Disk Array Enclosure (DAE) ............................................................................................... 12
Customer Replaceable Units (CRU) ................................................................................... 14
Storage Processor ........................................................................................................ 14
Lithium-Ion Battery Backup Unit (BBU) ......................................................................... 15
Power Supply ............................................................................................................... 16
Fan Modules................................................................................................................. 16
Dual-Inline Memory Module (DIMM) ............................................................................. 17
mSATA (Mini-Serial Advanced Technology Attachment) ................................................ 17
I/O module ................................................................................................................... 18
Hardware Configurations .................................................................................................. 18
Comparison of Drive Types ........................................................................................... 19
RAID Configurations ..................................................................................................... 20
Executive Summary
Year after year, information continues to grow at an astonishing rate. This growth is
transforming all aspects of businesses as they try to manage this ever increasing flow
of information and data. Consequently, increasing complexity is commonly seen in IT
departments because they continually require additional servers and applications to
serve their businesses needs. To satisfy those needs, storage systems in data center
environments need to be powerful enough to provide efficient access to businesses
data while ensuring protection and constant availability of that information.
Additionally, IT departments are constantly seeing demands for more capacity and
higher performance while being asked to maintain low costs in their data centers.
This makes it difficult to choose the right storage system for their environments.
These concerns are alleviated with the new EMC VNXe3200 storage system, a
unified storage solution that is useful for a wide range of organizations as well as
Remote Offices/Branch Offices (ROBO). This powerful storage system addresses the
challenges mentioned above and is designed to be simple such that IT generalists
with limited storage expertise are able to leverage the multitude of features that the
system provides. The major benefits of the VNXe3200 storage system includes:
High Availability VNXe3200 has many High Availability options available such
as hardware redundancy and RAID protection ensuring constant data flow to users
and administrators.
Performance and Efficiency By using the features of Multicore FAST Cache and
FAST VP, the VNXe3200 includes the ability to better leverage Flash and other
technologies to offer higher price/performance ratios without sacrificing any other
functionalities/capabilities.
Unified Snapshots Due to the truly unified infrastructure, the VNXe3200is able
to leverage the Unified Snapshots technology and take point-in-time snapshots of
block or file data.
Fibre Channel Support VNXe3200 systems supports the Fibre Channel protocol
which allows for seamless implementation into Fibre Channel network
infrastructures in customer environments.
Audience
This white paper is intended for IT planners, storage architects, administrators,
partners, EMC employees and any others involved in evaluating, acquiring, managing,
operating, or designing an EMC networked storage environment using VNXe3200
storage systems.
Terminology
Common Internet File System (CIFS) An access protocol that allows access to
files and folders from Windows hosts located on a network.
File System A storage resource that is associated with file access protocols. File
systems are accessed through Windows shares (CIFS protocol) or Linux/UNIX
shares (NFS protocol).
Fully Automated Storage Tiering for Virtual Pools (FAST VP) Advanced data
services that automatically relocates slices of LUNs or file systems data between
storage tiers with different performance and capacity capabilities to optimal
locations within that storage pool.
Link Aggregation A high-availability feature that allows Ethernet ports that are
associated with the same storage resources and have similar characteristics, to
combine into a single virtual device/link. The aggregated link may have a single IP
address or multiple IP addresses.
Logical Unit Number (LUN) Identifying number of a SCSI storage resource that
processes SCSI commands. The LUN is the last part of the SCSI address for a SCSI
object. The LUN is an ID for the logical unit, but the term often refers to the logical
unit itself.
MCx Multicore initiative (Multicore Cache, Multicore FAST Cache, Multicore RAID)
that delivers high performance and platform efficiency in VNXe3200 systems.
Multicore Cache VNXe3200 cache that optimizes Storage Processors DRAM and
cores usage to increase host write and read performance.
Multicore FAST Cache Large capacity secondary cache that uses enterprise Flash
drives. This secondary cache is meant for spikes in I/O to benefit applications
with unpredictable I/O profiles.
Multicore RAID MCx component that defines, manages, creates, and maintains
VNXe3200 RAID protection.
Network File System (NFS) An access protocol that allows users to access files
and folders from Linux/UNIX hosts located on a network.
Server Message Block (SMB) The underlying protocol used by CIFS to request
files, print, and communicate with a server over a network through Ethernet ports.
VNXe3200 storage systems support Microsofts Server Message Block (SMB) 3.0
for CIFS shares.
VNXe3200
The VNXe3200 storage system (Figure 2) is an ideal platform for businesses with
physical server infrastructures, as well as those making the move to server
virtualization for drive consolidation and greater efficiency. As an entry-level unified
storage system, the VNXe3200 provides high value for smaller user configurations
with lighter performance needs. It is designed to support smaller businesses and
applications for up to 1000 users. Also, the storage system has the increased
performance, scalability, and I/O expandability of VNX systems so it can be used with
larger configurations with higher performance demands. VNXe3200 systems are
available with a comprehensive set of features including increased capacity
utilization, data protection solutions, availability solutions, and advanced support
capabilities.
Hardware Overview
Quad-Core CPU
The VNXe3200 uses an Intel Xeon E5 Quad-Core CPU which runs at 2.2 GHz. The
VNXe3200 takes advantage of that processing power by leveraging its MCx
architecture to ensure that all the cores are optimally being used.
10
The back of the DPE (Figure 5) looks the same for both 25-drive and 12-drive DPE
configurations. The DPE includes an onboard 4-port 10 GbE Base T module, as seen in
Figure 6, for each SP which can auto-negotiate between 10 GbE/1 GbE/100 MbE/
10 MbE depending on the infrastructure. Also, there are two SAS ports allowing for
expansion of storage through DAEs. Both the management and service ports are 1
GbE ports which can auto-negotiate to 100 MbE/10 MbE. The management port is
where Unisphere (GUI) and SSH commands communicate with the system and if the
system is unavailable through the management port, the service port can be utilized
to directly connect using the Serial Over LAN feature. An I/O expansion module is
available to the VNXe3200 which adds Fibre Channel connectivity to the system.
11
Number
1
2
3
4
5
6
7
8
9
10
11
12
Component
Fan Module
Power Supply
10 Gb/s Ethernet Port
6 Gb/s SAS Port
SP Power LED
Unsafe to Remove LED
SP Internal Fault LED
SP Fault LED
Mini USB Port
I/O module (Fibre
Channel)
Management Port
Service Port
12
Number
1
2
3
4
5
6
7
8
Component
6 Gb/s SAS Primary
Port
6 Gb/s SAS Extension
Port
SAS Link LEDs
Back-End Bus LEDs
Power LED
Fault LED
Enclosure ID LED
SPS Port (Not Used)
13
Number
1
2
3
4
5
6
7
Component
6 Gb/s SAS Primary Port
6 Gb/s SAS Extension
Port
SAS Link LEDs
Back-End Bus LEDs
Enclosure ID LED
Fault LED
Power LED
14
15
Power Supply
Each DPE has two power supply modules (Figure 13), one for each SP. One of the two
power supplies must remain in service for the system to stay in an operating state.
16
17
I/O module
The VNXe3200 supports the option to add an additional I/O module for increased
connectivity. It is important to note that the Storage Processor configuration must
mirror each other. For example, if a Fibre Channel I/O module is installed in Storage
Processor A, then an identical Fibre Channel I/O module must be installed in Storage
Processor B.
There is currently one supported I/O module for the VNXe3200 which is a quad-port
8 Gb/s FC optical I/O module (Figure 17) that is used for frontend connectivity from
Storage Processors to host(s). It auto-negotiates to 2 Gb/s, 4 Gb/s, or 8 Gb/s, has an
8-lane PCI-E Gen-2 interface.
Hardware Configurations
Table 1 shows the hardware features available to the VNXe3200 storage system.
Table 1: VNXe3200 details
Model
VNXe3200
Number of SPs
Form factor
2U
1 (3.5) 12-drive*
or
1 (2.5) 25-drive*
Minimum/maximum number of
drives
6/50*
18
Model
VNXe3200
100 | 200 GB Flash
300 | 600 GB SAS
2 | 4 TB NL-SAS
12-drive (3.5")
or
25-drive (2.5")
Drive options
DPE options
CPU per system
48 GB
Back-end connection
6 Gb SAS
4 x 10 GbE
RAID options
1/0, 5, 6
500
16 TB
16 TB
438 TB
19
and images) and for users with applications that do not have strict performance
requirements.
For a list of supported drive types, please see Appendix A: Drive Support.
RAID Configurations
With a VNXe3200, the drives can be configured into storage pools which are
repositories of homogeneous or heterogeneous grouping of physical disks from which
storage resources can be provisioned. Storage pools are easy-to-use, economical
alternative to traditional storage system provisioning.
Each tier in a storage pool has a RAID type with the desired capacity and drive
characteristics. A FAST VP license is required to support the creation of multi-tier
storage pools. The supported RAID levels are 5, 6, and 1/0.
Note: Once a specific tier in a storage pool is configured, the RAID type cannot change
for that tier.
RAID 5 is best suited for transactional processing, and is often used for generalpurpose storage, relational databases, and enterprise systems. This RAID level
provides fairly low cost per gigabyte while still retaining redundancy and high
availability. RAID 5 stripes data at a block-level across several disks and distributes
parity among the disks. However, it is important to note that a failed disk reduces
storage performance and should be replaced immediately. However, data loss will not
occur as a result of a single disk failure.
RAID 6 is appropriate for the same types of applications as RAID 5, but more useful in
situations where providing increased fault tolerance is important. RAID 6 is similar to
RAID 5, but it uses a double parity that is distributed across different disks, so it
offers higher fault tolerance. In this RAID configuration, storage pools can continue to
operate with up to two failed disks. Also, VNXe3200 systems have the ability to
perform a parallel rebuild in case of two faulted disks, decreasing the time in a
faulted state.
RAID 1/0 is a stripe of mirrors so only half the total capacity of the drives is usable.
This provides high performance and reliability, but at high cost because of lower
usable capacity per disk. RAID 1/0 may be more appropriate for applications with fast
or high processing requirements as well as write-intensive workloads, such as
enterprise servers and moderate-sized database systems. RAID 1/0 in VNXe3200
systems requires a minimum of two physical disks, where a mirrored set provides
fault tolerance. Note that a two disk RAID 1/0 configuration does not use striping.
Lastly, a faulted disk must be replaced immediately because this configuration
cannot handle the loss of more than one disk in the same mirrored pair.
20
Software Overview
This section discusses the software features available in VNXe3200 systems.
Initial Configuration
When configuring a VNXe3200 system for the first time, the management interface
must be configured with an IP address. An IPv4 or IPv6 address can be assigned
dynamically via DHCP or with a static IP address using the Connection Utility.
A DHCP server on the network will assign an IPv4 or IPv6 address to the VNXe3200
systems management interface automatically upon the initial boot up of the storage
system.
If there is no DHCP server, you can use the Connection Utility to assign a static IP
address for managing the VNXe3200 storage system. This utility can be downloaded
directly from the VNXe3200 product support page on the EMC Online Support
website. When running the Connection Utility from a computer on the same subnet as
the storage system, the program automatically discovers any unconfigured VNXe3200
systems. If the Connection Utility computer cannot be placed on the same subnet, the
configuration can be saved to a USB drive and transferred by plugging it into the mini
USB port on the back of the DPE.
After configuring an IP address for the VNXe3200 management interface, you can
access Unisphere by navigating to the assigned IP address in any web browser.
When connecting to a VNXe3200 storage system for the first time, the Unisphere
Configuration Wizard starts automatically. This wizard runs you through a series of
steps to seamlessly configure the VNXe3200 system and create storage resources in
a short amount of time. The following parameters can be set using the Configuration
Wizard in Unisphere:
Unisphere licenses
21
Unisphere
Unisphere is a graphical, management interface with a web-familiar look and feel
making it an ideal monitoring and management tool for VNXe3200 storage systems.
First and foremost, it provides tools for creating, configuring, and managing storage
resources including file systems, LUNs, and VMware datastores. To make use of those
storage resources, access needs to be given to users, applications, and/or hosts. This
is done through different supported protocols including iSCSI, Fibre Channel, CIFS,
and NFS. Once users start utilizing those storage resources, the files and data stored
on the storage system needs to be protected for situations such as retrieving
accidentally deleted files. To do this, an automated scheduler feature is available for
users to create point-in-time snapshots of stored data ensuring protection. This
makes recovering lost files and data from recorded snapshot images quick and easy.
Users also want to make sure their storage systems stay in a healthy operating state,
so knowing where and when any potential issues can occur is important. Unisphere
enables users to monitor storage operations and the system status through a detailed
graphical reporting service, which can pinpoint issues to specific components of the
storage system. Unisphere also enables monitoring of storage system performance
through graphs allowing technical users to see a deeper view of system utilization.
Technical support is a must with any software product and Unisphere provides direct
support through ecosystem resources like online documentation, white papers, FAQs,
how-to videos, online discussion forums, and chat sessions. Lastly, any software
code updates on the VNXe3200 can be easily done through a non-disruptive upgrade
(NDU) process directly from Unisphere.
You can use Unisphere to create, configure, and monitor the following types of
storage resources:
LUNs Provides generic block-level storage, in the form of LUNs, for hosts and
applications to access using iSCSI or Fibre Channel protocols.
File System File-level storage for hosts. File systems can be accessed
through CIFS or NFS protocols.
22
The Dashboard page in Unisphere (Figure 18) provides a convenient starting point for
various operations.
For additional information about Unisphere for VNXe3200 storage systems, please
see the paper titled EMC Unisphere for the VNXe3200: Next-Generation Storage
Management.
23
MCx
The VNXe3200 brings the powerful MCx feature to its architecture. MCx is a
combination of Multicore Cache, Multicore RAID, and Multicore FAST Cache features,
which enables the array to fully leverage the Intel multicore CPU architecture. MCx
was designed from the ground up to ensure optimum performance at high scale with
little to no customer intervention, all in an adaptive architecture. It delivers core
functionality improvements that make the new VNXe3200 platform robust, reliable,
and easy to use. The MCx architecture is a true enterprise-ready storage platform and
comes free with the VNXe3200.
Multicore Cache
Multicore Cache is an intelligent, adaptive caching mechanism that removes the need
to manually assign space for read and write cache. Also, there is no management
overhead in ensuring the cache is working in the most effective manner regardless of
the I/O mix coming into the system. Cache allocation occurs on the fly for both reads
and writes, while dynamically adjusting and constantly optimizing regardless of
changing workloads. SP Cache management is fully automated and implements
highly advanced caching algorithms to optimally handle varying sequential/random
and read/write workloads. The system can also throttle write arrival rates which
lessens the need for forced flushes. A forced flush is when the contents of write cache
are written to the back-end disks to free up space for more writes.
Multicore RAID
Multicore RAID enhances handling of I/O to back end storage (HDDs and SSDs).
Modularization of back end data management processing allows for a seamless scale
across all processors.
In addition to this, the VNXe3200 delivers improved flexibility, ease of use,
performance, and reliability with new features like Permanent Sparing, Portable
drives, and RAID 6 Parallel Rebuild.
Permanent Sparing
Instead of having to define disks as permanent hot spares, the sparing function is
now significantly more flexible. Any unassigned drive in the system is available to be
used as a spare drive. The system automatically sets a sparing policy of 1 spare per
30 drives. When a drive is used in a sparing function, the particular drive becomes a
permanent member of the RAID configuration. There is no need to copy back
(equalize) the data when the faulted drive is replaced, thereby reducing the exposure
and performance overhead of sparing operations.
Portable drives
Drives can be physically removed from its slot and relocated to another slot or disk
shelf in the same array. This feature can be used to rebalance the buses in an active
system. If the disk is not replaced in the same VNXe3200 within five minutes, a
permanent sparing operation will occur. It is also possible to remove entire RAID
24
configurations for longer periods (should there be a need to store data offsite). Drives
can be moved within the same system, but cannot be moved from one storage system
to another since they will be zeroed.
RAID 6 Parallel Rebuild
Multicore RAID supports the rebuild of two drives in parallel RAID 6 configurations. For
example, if one disk fails, a rebuild process will occur on a suitable spare drive after
five minutes. If a second disk fails during the rebuild, the second rebuild process will
start on another spare, but at the same place where the first rebuild is occurring. For
example, 50 percent of the first rebuild done, the second will start at the 50 percent
mark. After the second spare rebuilds to the end of the drive, it starts back at the top
of the drive until it finishes the rebuild. This allows for improved resiliency and
protection, as the RAID 6 configuration is reconstructed and restored quicker.
Multicore FAST Cache
Multicore FAST Cache remains an industry-leading feature for EMC storage systems.
The VNXe3200 uses this feature to extend the systems read/write cache to ensure
that unpredictable I/O spikes are serviced at Flash speeds, which benefits all
applications (both block and file). FAST Cache complements FAST VP, as it works at a
more granular level, by copying 64 KB chunks onto Flash drives depending upon the
I/O characteristics.
Three hits to the same 64 KB chunk of data causes the system to promote that data to
FAST Cache. FAST Cache also works in quicker cycles than FAST VP; it has a response
time in the order of milliseconds to microseconds. Since FAST Cache reacts well to I/O
spikes and maintains access levels, it acts as an extension to the onboard memory.
FAST Cache is most appropriate for workloads with a high locality of reference. For
example, applications that access a small area of storage with very high frequency,
such as database indices and reference tables.
For more information on Multicore FAST Cache, please see the white paper titled
Introduction to the EMC VNXe3200 FAST Suite.
25
FAST VP
Fully Automated Storage Tiering for Virtual Pools (FAST VP) feature is introduced in the
VNXe3200, and is available for both block and file data. FAST VP optimizes storage
utilization by automatically moving data between and within storage tiers (Extreme
performance, Performance, and Capacity).
26
Storage Provisioning
The VNXe3200 includes both block and file connectivity. For the block connectivity, a
user has the option to use iSCSI interfaces or Fibre Channel interfaces for host access.
For file connectivity, a user has the option to configure NAS servers which is used to
access CIFS or NFS shares. If a user wants to create VMware datastores, they can
provision datastores via NFS using an available NFS-enabled NAS server or VMFS
through iSCSI or Fibre Channel interfaces. See Table 2 for the different storage types
available to VNXe3200 storage systems.
Table 2: VNXe storage details
Storage Type
Hosts
Managed Objects
Interface
Linux/UNIX hosts
File systems
Windows hosts
File systems
LUNs
Windows or Linux/UNIX
hosts that have iSCSI or
FC connections
LUNs
iSCSI/FC interfaces
NFS datastores
VMware datastores
VMware hosts
VMFS datastores
iSCSI/FC interfaces
File systems
File Systems
File systems and shares provide file-level storage resources that hosts can access
over network connections. A file system is a file-based storage resource that is
accessed via file access protocols (CIFS or NFS). A share is an access point through
which hosts can access file systems.
When file-level storage is enabled on the system, a NAS server maintains and
manages the configured file systems, and transfers data to/from hosts and shares.
The type of access to a VNXe3200 share varies depending on whether the share is a
CIFS share or an NFS share.
When creating a NAS server on a VNXe3200 system, the CIFS and/or NFS protocol can
be enabled. However, even if both protocols are enabled on the NAS server, only
single-protocol shares can be made (Figure 20). For example, if the CIFS protocol is
chosen when creating a file system, only CIFS shares can be created for that file
system. To create NFS shares, a separate file system would have to be created.
27
28
VMware Datastores
Unisphere enables a user to create storage resources optimized for use by VMware
vCenter servers and ESXi hosts. VMware datastores are associated with a type of
storage access (VMFS or NFS).
The Hosts page in Unisphere allows a user to discover and add VMware ESXi hosts
and vCenter server information to Unisphere. Once the VMware host information is
imported into Unisphere, information about virtual machines deployed on those
servers is displayed.
A user can configure VMware datastores by using the VMware Storage Wizard, which
configures datastore settings.
Virtual Machine File System (VMFS) Datastores
VMFS datastores require iSCSI initiators or Fibre Channel connections on the storage
system to be created as well as have an ESXi host connected to the array. After the
creation of a VMFS datastore, the capacity can be increased, but not reduced. When
VMFS datastores are created and an ESXi host is given access, the new storage is
automatically scanned and made available to the ESXi host. The following host
access configurations can be set for VMFS datastores: LUN, Snapshot, LUN and
Snapshot, and No Access.
Network File System (NFS) Datastores
NFS Datastores require a NAS server to be configured beforehand. Linux/UNIX hosts
can then be given host access to the NFS datastores. The following host access
configurations can be set for NFS datastores: Read-Only, Read/Write, Read/Write,
allow Root, and No Access.
Storage Pool Characteristics
VNXe3200 storage can be provisioned from storage pools. A storage pool is a
homogeneous (Figure 22) or heterogeneous (Figure 23) grouping of physical disks
with each tier being disks of similar type and speed. A homogeneous storage pool is a
single tier pool using one type of disk. A heterogeneous storage pool consists of
multiple tiers using different types of disks.
Heterogeneous pools are available to VNXe3200 systems and require a FAST VP
license to use. The three tiers available in the VNXe3200 are extreme performance,
performance, and capacity. The RAID configuration can be customized for each tier in
a storage pool. Once a RAID configuration has been set, it cannot be changed. Also,
there is a built in hot spare policy which is 1 spare drive per 30 drives. This policy is
not user configurable.
29
Extreme Performance
Performance
Capacity
30
If a license for FAST VP is not available, a Quick Start Mode will be available in the
Storage Pool Wizard. This Quick Start Mode will automatically create homogeneous
storage pools using a best fit algorithm for all of the available disks on the system
while incorporating the built in hot spare policy. Quick Start Mode will only configure
RAID 5 (Flash and SAS) or RAID 6 (NL-SAS). If automatic pool creation is not wanted,
users can create customizable homogeneous storage pools.
When creating a custom storage pool, users have the option to specify stripe width or
choose maximum capacity which will use a best fit algorithm for the available disks.
The allowed stripe widths for each RAID type for VNXe3200 systems are as follows for
VNXe3200 systems:
Anti-Virus Protection
LDAP Integration
Link Aggregation
Thin Provisioning
Unified Snapshots
User Management
31
Anti-Virus Protection
VNXe3200 storage systems support third-party anti-virus servers that perform virus
scans and reports back to the VNXe3200. For example, when a client creates, moves,
or modifies a file, the storage system invokes the anti-virus server to scan the file for
known viruses. If the file does not contain a virus, it is written to the VNXe3200. If the
file is infected, corrective action is taken as defined by the anti-virus server.
VNXe3200 supports the following anti-virus servers:
McAfee VirusScan
Sophos Anti-Virus
Kaspersky Anti-Virus
32
still be displayed, but they will share the same file data. The one instance of the
shared file is also compressed, providing further space savings.
During deduplication, each deduplication-enabled file system is scanned for files that
match specific criteria, such as a particular file type or a modification time older than
a certain date. When a file is found that matches the specified criteria, the file data is
deduplicated and compressed if appropriate. Since file metadata is not affected by
deduplication, different instances of the file can have different names, security
attributes, or timestamps.
For more information on File deduplication and compression, please see the white
paper titled EMC VNXe3200 File Deduplication & Compression.
File-Level Retention (FLR) Enterprise
File-Level Retention (FLR) provides a way to set file-based permissions on a file
system to limit write access to the files for a specific period of time. FLR ensures the
integrity of data during that period by creating an unalterable set of files and
directories. FLR can only be enabled when initially creating a file system. After a file
system is created and enabled with FLR, protection is applied on a per-file basis. Only
FLR-Enterprise (FLR-E) is available for VNXe3200 systems. For FLR-E, an administrator
with appropriate authorization can delete an FLR-E file system even if it has protected
files, but he/she cannot modify or delete individual protected files.
A file in an FLR-enabled file system is always in one of the following four states:
Not Locked A file that is not locked is treated exactly as a file in a file system
that is not enabled for FLR. The file can be renamed, modified, or deleted.
For more information about File Level Retention, please see the white paper titled,
Managing an FLR-Enabled NAS Environment with the EMC File-Level Retention Toolkit.
LDAP Integration
Lightweight Directory Access Protocol (LDAP) is an application protocol for querying
and modifying directory services running on TCP/IP networks. LDAP provides a
management service for network authentication and authorization operations by
centralizing user and group management across the network. Integrating Unisphere
users into an existing LDAP environment provides a way to control management
access to the VNXe3200 system based on established user and group accounts
within the LDAP directory. In Unisphere, LDAP settings can be configured under the
Directory Services tab of the Manage Administration page.
33
Note: LDAP settings are used only for controlling access to Unisphere, and not for
controlling access to VNXe3200 storage resources.
Link Aggregation
VNXe3200 systems support link aggregation allowing up to four Ethernet ports to be
combined into one logical link. The cabling on Storage Processor A (SPA) must be
identical to the cabling on Storage Processor B (SPB). Link aggregation provides the
following advantages:
High availability of network paths to and from VNXe3200 If one physical port
of a link aggregated port group fails, the VNXe3200 system does not lose
connectivity.
For example, if two physical ports are combined into one logical port and one of the
links fails, the other link carries the network traffic without disrupting host access.
Once the link returns, network traffic runs on both healthy links again.
In order to configure link aggregation, both linked ports must be connected to the
same switch or the same logical switch if the switches support stack interconnects
that allow for cross-stack link aggregation. The switch must be configured to use link
aggregation using the Link Aggregation Control Protocol (LACP).
LACP allows the combination of any Ethernet port to any other Ethernet port on the
same card.
Thin Provisioning
Thin provisioning is the ability to present a server with more capacity than is actually
allocated within the storage system (Figure 6).
When a user enables thin provisioning for a storage resource, the amount of storage
requested is not allocated to the resource immediately. Instead, the system allocates
a smaller quantity of storage, known as the initial allocation, for file resources and no
initial allocation for block resources. When the amount of storage consumed within
the storage resource approaches the limit of the current allocation, the system
automatically allocates additional storage space to the storage resource from the
storage pool. Thin provisioning is supported on all storage types in a VNXe3200
storage system and is enabled by default via a checkbox in storage resource creation
wizards.
VNXe3200 systems allows multiple storage resources to subscribe to a common
storage pool and, with thin provisioning, the system allocates only a portion of the
physical capacity requested by each storage resource (Figure 24). The remaining
storage is available in the pool for use by other storage resources.
34
35
data over the service port on the SP to establish an SOL session. Once the session is
started, regular commands can be run on the system like in SSH.
Note: Files cannot be transferred over the service port.
For more information on Serial Over LAN, please see the technical note EMC IPMI Tool
For VNXe3200 systems, EMC introduced the Unified Snapshots feature which creates
point-in-time views of data for both block and file storage resources. Unified
Snapshots are based on Redirect on Write (ROW) technology which means, after a
snapshot is created, new writes to the storage resource are redirected and written to a
new location in the same storage pool. Therefore, a snapshot does not consume
space from the storage pool until new data is written to the storage resource or to the
snapshot itself.
Unified Snapshots are easy to manage and all required operations can be done using
Unisphere. Unified Snapshots also provide the ability to schedule snapshots. This
means users can automatically protect their data periodically and can recover that
data in the event of a deletion or corruption.
For more information on Unified Snapshots, please see the white paper titled EMC
VNXe3200 Unified Snapshots.
Unisphere CLI Support
The Unisphere CLI in the VNXe3200 system enables a user to perform scripting for
commonly performed tasks. The VNXe3200 management IP address can be used in
Unisphere CLI to execute commands on that system.
To use Unisphere CLI, the Unisphere CLI client must be installed on a host machine.
The same client can be used to manage multiple VNXe3200 storage systems.
For more information about the Unisphere CLI format and options, please see the
VNXe Unisphere CLI User Guide on EMC Online Support.
User Management
VNXe3200 storage systems provide tools for creating user accounts for managers and
administrators who configure and monitor VNXe3200 systems. When users access
Unisphere, they are prompted to log in with account-based credentials before they
can access the system. Unisphere user accounts combine a unique username and
password with a specific role for each account. The role determines the types of
actions that the user can perform after logging in.
The following user roles are available in Unisphere:
Operator This role can view Unisphere system and storage status
information, but cannot change system settings.
36
Storage administrator This role can view VNXe3200 storage system data,
edit Unisphere settings, use Unisphere tools, create/delete storage resources,
and create/delete host configurations. However, this role cannot add user
accounts, perform initial configuration of the system, modify network settings,
modify NAS Server configuration, or upgrade system software.
Administrator This role can perform all tasks accessible through Unisphere.
Power supplies Each DPE and DAE has two power supplies. They can
individually continue to run with one failed power supply in the enclosure. The
SP also contains a battery backup unit (BBU) that provides enough backup
power to write cached data to non-volatile media in case of power interruption.
37
Network links For NAS servers, if a network link fails, link aggregation, if
configured, provides an alternate path if it is enabled.
RAID protection The data on the disks is protected with RAID 5, RAID 6, or
RAID 1/0. With RAID 5, data is still available after a single physical drive
failure. With RAID 6, data is available in the event of a double disk failure. RAID
1/0 uses mirroring for availability and striping for performance. If a disk in the
mirrored pair fails, the mirror provides continued access to data.
For additional information about HA in VNXe3200 storage systems, please see the
white paper titled EMC VNXe3200 High Availability.
38
For additional information about VSI Storage Management, please see the product
guide titled VSI for VMware vSphere: Unified Storage Management on EMC Online
Support.
NAS
o
o
o
Block
o
o
o
o
Fast Copy
Extended Statistics
Reserve Space
Fully Copy or Hardware-Assisted Move
Block Zero or Hardware-Assisted Zeroing
Atomic Test and Set (ATS) or Hardware-Assisted Locking
Thin Provisioning (Dead Space Reclamation)
39
Remote Monitoring
The status of a VNXe3200 system can be managed remotely by EMC personnel or customers
through ESRS. ESRS is useful for troubleshooting purposes if the VNXe3200 experiences
issues. With ESRS, EMC technical support can easily take a look at the system enabling faster
time to resolution.
How to videos Videos to learn about the storage system. For example,
instructions on how to replace a failed component.
Search EMC Support Articles, white papers, and other information regarding
known issues and solutions related to system installation, configuration, and
operation.
Live Chat A way to quickly contact and chat with support personnel who can
help a user in real time. Note that this option is available only to VNXe3200
systems that are under a maintenance contract.
40
Product Support Page A way to access all the system support needs such as
information about maintaining hardware as well as steps to install and license
the system.
Customer Replaceable Parts A place to order or return a part for the storage
system.
Also, most wizards and screens in Unisphere display Help icons that lead you to
specific pages in the Unisphere Online Help.
Conclusion
The VNXe3200 storage system provides a consolidated platform for provisioning,
managing, and monitoring data storage for a wide range of organizations and
networks. Each storage system provides highly-efficient iSCSI/FC block storage and
highly-accessible file-based storage that enables a user to work with files and data
over network connections. Because these storage systems support SAN (iSCSI and
FC) and NAS (network-attached storage) storage access, they provide storage
resources for clients and hosts running a variety of operating systems and
applications. The VNXe3200 system is a truly unified data storage system.
Since VNXe3200 software automatically implements best practices when configuring
storage, a user does not need detailed knowledge about storage and application
technologies. Along with powerful features like MCx and FAST VP, VNXe3200 systems
are designed to be easily installed, configured, and maintained by an IT generalist.
Comprehensive, easy-to-understand guidance is also available through the
VNXe3200 support ecosystem.
From a hardware point of view, a user can start with a small configuration based on
current requirements, and easily scale up as requirements change. The ability to
easily implement advanced storage features with application wizards such as File
Deduplication and Compression, Thin Provisioning, and Unified Snapshots in a
solution that is also integrated with VMware vCenter makes VNXe3200 storage
systems unique in the market place.
IT departments can greatly benefit from using the VNXe3200 as it helps businesses
handle exponentially growing storage capacity demands while providing reliable
performance, availability, and support.
41
25 Drive
DAE
Drive Type
Capacity
Size
100GB
2.5
200GB
2.5
100GB
2.5
200GB
2.5
300GB
2.5
600GB
2.5
300GB
3.5
600GB
3.5
2TB
3.5
4TB
3.5
42
References
The following documents can be found on EMC Online Support VNXe3200 Product
Page:
43