EMC RecoverPoint Technical Review

Report
EMC RECOVERPOINT
TECHNICAL REVIEW
Network-based intelligent data
protection
© Copyright 2010 EMC Corporation. All rights reserved.
1
Business Continuity Considerations
• What are your company’s most
critical processes and data needs?
• How much data can you afford to
lose?
• How quickly do you need to restore
your critical processes?
• How vulnerable are your operations
to disasters?
Answer these questions with the right
solution to mitigate these risks
© Copyright 2010 EMC Corporation. All rights reserved.
2
Overview and Systems
Architecture
© Copyright 2010 EMC Corporation. All rights reserved.
3
Family Overview
Replication independent of host and storage platforms
SAN
RecoverPoint
FC/WAN
RecoverPoint
SAN
• Replicate data
bi-directionally across
different storage arrays
from the same vendor—
or different vendors
– RecoverPoint/SE supports a
single CLARiiON or unified array
per-site
– RecoverPoint supports multiple
EMC and non-EMC storage
arrays
• Creates independent local
and/or remote copies of data
• Automates production
rebuild/restart for testing,
business continuity, and
disaster recovery
© Copyright 2010 EMC Corporation. All rights reserved.
4
Family Details
RecoverPoint—One solution for data
protection, replication, and disaster
recovery in mixed storage environments
• Heterogeneous storage array support
Common functionality
 Continuous data protection (CDP) for one set of LUNs
in one SAN
 Continuous remote replication (CRR) between LUNs in
two SANs
• Integrated with intelligent fabric from Brocade
and Cisco
 Continuous local and remote (CLR) data protection
that combines CDP and CRR
• Supports up to 600 TB of production data
 Flexible point-in-time recovery without production
impact
RecoverPoint/SE (for EMC CLARiiON)—
One solution for data protection,
replication, and disaster recovery for
CLARiiON
environments
• Supports Microsoft Windows host-based and
CLARiiON array-based splitting only
 Heterogeneous operating system support
 Wizards automate production rebuild/restart from any
point in time
 Snapshot consolidation enables application-based
recovery point objective (RPO)/recovery time objective
(RTO); near-instant recovery reduces RTO
 Application integration for Microsoft Exchange and
SQL, other applications and databases
• Single CLARiiON or Celerra Unified per site
 Integrated consistency groups support federated
servers and storage
• Supports up to 150 TB of production data
 Synchronous or asynchronous local and/or remote
replication with optional dynamic selection policies
© Copyright 2010 EMC Corporation. All rights reserved.
5
Bi-directional Local and Remote
Replication
PRODUCTION SITE
Application
servers
RecoverPoint
appliance
Local
copy
OPTIONAL DISASTER RECOVERY SITE
RecoverPoint bi-directional
replication/recovery
Remote
copy
Standby
servers
RecoverPoint
appliance
Production and
local journals
Prod
LUN
s
SAN
Host-based splitter
Storage
arrays
FC/WAN
Remote
journal
SAN
Storage
arrays
Fabric-based splitter
CLARiiON-based
splitter
© Copyright 2010 EMC Corporation. All rights reserved.
6
Components and Topology
• RecoverPoint appliance (RPA)
Application servers
– Connects to LAN and FC/WAN for replication
management and replicating data over FC/IP
– Connects to Layer 2 FC SAN for storage
access
– Sold and supported with RecoverPoint
– Installed as two to eight appliances per side
RecoverPoint nodes
Layer 2
Fibre Channel
SAN
FC/WA
N
• Application servers
– Accesses data that needs to be replicated
EMC IBM
HDS LSI
HP
• Heterogeneous storage
– Source or Target for replicated data
Heterogeneous storage
• RecoverPoint appliance cluster
– A group of inter-linked RPAs, working
together closely, to provide replication
services
– RPAs in a RecoverPoint cluster are called
nodes
© Copyright 2010 EMC Corporation. All rights reserved.
7
Performance and Scalability
• Each RecoverPoint cluster
– Supports up to 128 consistency groups
– Replicates up to 2,048 LUNs
• Each RecoverPoint appliance (RPA)
– Has a steady state rate of 75 MB/s
– Eight RPAs with one consistency group each
supports an aggregated write rate of 600 MB/s
– Four RPAs (maximum number for a distributed
consistency group) will provide a distributed
consistency group with a >240 MB/s steadystate rate for asynchronous replication
© Copyright 2010 EMC Corporation. All rights reserved.
8
Architecture and Packaging
 RecoverPoint appliance (RPA) based
on a Dell R610 1u server

–
–
–
–

Hosts RecoverPoint software
Two to eight appliances per side
Rack in EMC Titan rack or third-party rack
Cannot be racked in EMC Connectrix rack
 Management application
– Any server with SSH and IP connectivity to a
RecoverPoint appliance
– Java-based RecoverPoint GUI for Windows
– Installation and configuration management
– Monitoring and recovery management

 Storage
CLARiiON
Symmetrix
© Copyright 2010 EMC Corporation. All rights reserved.
Third party
– RecoverPoint/SE: one CLARiiON array per
side
– RecoverPoint: multiple CLARiiON,
Symmetrix, and third-party arrays per side
9
RecoverPoint Appliance High Availability
• Platform
Hosts
– No single point of failure
• Function
Layer 2 SAN
(A/B fabric)
Storage
arrays
– Automatic RecoverPoint
appliance failover
– External repository for status
ensures no data loss during
appliance failover
• Service
– Non-disruptive software updates
– Remote maintenance, call-home,
and automatic diagnostics
– Hardware monitoring
© Copyright 2010 EMC Corporation. All rights reserved.
10
Distributed Consistency Group
Architecture
• Handles more write activity (‘throughput’) than a single RPA
• Spread LUN write activity across multiple RPAs
• Avoids any bottlenecks caused by a single WAN connection
TRIBUTED CONSISTENCY GROUP OPERATION DURING ASYNCHRONOUS REPLICAT
Application
servers
Replica
volumes
RecoverPoint appliances:
Primary and three secondary
SAN/WAN
Production volumes
© Copyright 2010 EMC Corporation. All rights reserved.
11
Distributed Consistency Groups
Sample display
© Copyright 2010 EMC Corporation. All rights reserved.
12
Write Splitters
• Host-based, fabric-based, and CLARiiON-based write
splitters
• Intercepts writes from the host and splits each write
• Managed through RecoverPoint; installed and upgraded
independently of RecoverPoint
Splitter Type
How Deployed
Overhead
Host-based
In I/O stack just above the multi-path
software
Adds write traffic at the HBA; no other
impact
Fabric-based
In intelligent storage services hardware
on a Brocade- or Cisco-based switch
Operates at wire speeds; no impact
CLARiiONbased
In FLARE operating system; active in
both storage processors
No impact
RecoverPoint/SE only supports a Windows host-based write splitter and the CLARiiON-based write splitter.
© Copyright 2010 EMC Corporation. All rights reserved.
13
Host-Based Write Splitters
• Supported on AIX*,
Solaris*, and Windows
HOST (WINDOWS)
Kernel
File System
User Space
Communication
with the RPA
over FC
Partition Manager
Watchdog Service
Splitter User Space
Monitor the
splitter process
HIC/HLR
Kshsplit.sys
Various
application
processes
.
.
.
.
.
.
Multipath Driver
Disk Class Driver
(e.g., Disk.sys)
HBA Miniport/
Storport Driver
HBA
HBA
© Copyright 2010 EMC Corporation. All rights reserved.
Listen for Log
Request,
collect and
transfer to
RPA over IP
• Lightweight code that
installs on each server that
accesses a replicated LUN
• Splits or ‘Duplicates’ writes
to production LUNs and
sends a copy to the RPA
*AIX and Solaris host-based write splitters are
not supported by RecoverPoint/SE
NIC
14
Fabric-Based Write Splitters
• RecoverPoint leverages
intelligent fabric services
Applications
Copy of write
(split write)
• Supports Brocade Fabric
Application Platform
Appliance
Not in primary data path
APIs
Original write
Target
• Supports Cisco SANTap
• Supports EMC and non-EMC
Storage arrays
• Full support for VMware,
including VMFS
Not supported by RecoverPoint/SE
© Copyright 2010 EMC Corporation. All rights reserved.
15
Brocade Fabric-Splitter Platforms
• Department and Director versions
– Connectrix AP-7600B for department
– PB-48K-AP4-18 switching module for
Connectrix B-series ED-48000B and ED-DCXB directors
Connectrix AP-7600B
PB-48K-AP4-18
• Features
– High-performance storage processor
– 16 4 Gb/s ports
– 2 Gigabit Ethernet ports
• No RecoverPoint or Brocade license
required
Not supported by RecoverPoint/SE
© Copyright 2010 EMC Corporation. All rights reserved.
16
Cisco Fabric-Splitter Platforms
Connectrix MDS-9000 Series
with MDS 18/4 Multi-Services blade
MDS 18/4
Multi-Services Blade
Connectrix MDS-9222i
Native support or with MDS 18/4 Multi-Services blade
• MDS-9000 uses the 18/4
blade
• 9222i can use the 18/4 blade
and has native fabric splitting
support
• Both MDS and 9222i require
a Cisco Storage Services
Enabler (SSE) license, no
RecoverPoint license required
Not supported by RecoverPoint/SE
© Copyright 2010 EMC Corporation. All rights reserved.
17
CLARiiON-Based Write Splitter
DISASTER RECOVERY • RecoverPoint splitter
Windows Solaris Linux VMware
included at no charge with
FLARE 26 and later
PRODUCTION
Windows Solaris
Linux
VMware
RecoverPoint
RecoverPoint
SAN/WAN
SAN
SAN
• Supports CLARiiON CX3
and CX4 arrays
• Supports Celerra NS arrays
(Celerra Unified)
PRODUCTION A
• Supports open systems
CLARiiON splitter runs on each
CLARiiON; no host or fabric agent
required—
can share between clusters
RecoverPoint
SAN
• Can be mixed with other
RecoverPoint splitters
RecoverPoint
FC/WAN
SAN
• Easiest to use and
configure
• Can be shared by up to four
RecoverPoint clusters
VMware
Solaris
© Copyright 2010 EMC Corporation. All rights reserved.
VMware
Solaris
18
Splitter guidelines
• Different splitter types can be intermixed
Splitter
Hostbased
Brocade CiscoSANTap
CLARiiO
N
Host-based
Yes
Yes
Yes
Yes
Brocade
Yes
Yes
No
Yes
CiscoSANTap
Yes
No
Yes
Yes
• Multiple
supportedYes
CLARiiONinstances
Yesof a splitter
Yes type isYes
• CLARiiON-based write splitter can be shared
• The Cisco-SANTap Intelligent-Fabric splitter can be shared with an
RPQ
• Multiple versions of splitters are supported
© Copyright 2010 EMC Corporation. All rights reserved.
19
Splitter Support by OS
OS/Splitter Type
Host-based
Brocade
Cisco
CLARiiO
N
AIX
5.2, 5.3
Yes
Yes
Yes
HP-UX
No
Yes
Yes
Yes
Hyper-V
Hyper-V and Hyper-V
R2
Yes
Yes
Yes
Linux
No
Yes
Yes
Yes
Solaris
10, 11
Yes
Yes
Yes
VMware VM with
Windows
RDM/P only
Yes
Yes
Yes
VMware VM with other
No
Yes
Yes
Yes
Windows
32 and 64 bit, Server
2003, 2008, 2008 R2
Yes
Yes
Yes
Other OS
No
See ESM
See ESM
See ESM
© Copyright 2010 EMC Corporation. All rights reserved.
20
Stretched CDP over Fibre Channel
PRIMARY SITE
REMOTE BUNKER SITE
Application Database File and
servers
servers print servers
Application Database File and
servers
servers print servers
Management
IP WAN
RecoverPoint
Fibre
Channel
SAN
Production
volumes
Production
journal
 Supports synchronous Stretched CDP
CDP replication to
secondary site (such
as a bunker site)
 Enables survival of
primary site outage
 Provides synchronous
replication across
distance
 Remote applications
can access any pointin-time image as
read/write without
impacting production
SAN
 All RecoverPoint
resources installed at
remote site
 Dual fabric extended
to remote site
 Servers and splitters
must be at both sites
 Reversing replication
by promoting CDP
replica to production is
not supported
Replica
volumes
Replica
journal
Not supported by RecoverPoint/SE
© Copyright 2010 EMC Corporation. All rights reserved.
21
Cascaded Replication
• CLR replication for a three-site disaster restart solution
–
–
–
–
–
Replicates data from a primary site to a secondary site
Also replicates data from a primary site to a tertiary site
Improves RPO and RTO at tertiary site over standard CRR
Provides disaster restart by converting secondary site into production
Provides disaster recovery from the tertiary site
• Requires stretched fabric between primary and secondary
• Uses secondary site as bunker for replication data
• Retains recoverability at the tertiary site
– Survive a regional disaster that incapacitates both the primary and bunker
site
– Protects latest and all earlier consistent point-in-time images
Not supported by RecoverPoint/SE
© Copyright 2010 EMC Corporation. All rights reserved.
22
Cascaded Replication Topology
Local and remote data protection minimizes data loss
PRIMARY SITE
SECONDARY SITE
TERTIARY SITE
WAN
(for admin
functions)
WAN/FC
Fibre
Channel
SAN
SAN
SAN
Production
volumes
CDP replica
volumes
CRR replica
volumes
Production
CDP journal
CDP
journal
CRR
journal
Near-Synchronous
or Synchronous*
Synchronous over Fibre Channel limited by distance*
Asynchronous over Fibre Channel or WAN not limited by distance*
* Refer to the EMC Support Matrix for the maximum distance for your configuration.
© Copyright 2010 EMC Corporation. All rights reserved.
Not supported by RecoverPoint/SE
23
Cluster Enabler 4.0 for RecoverPoint and
RecoverPoint/SE
RecoverPoint
Fibre
Channel/
WAN
RecoverPoint
• Supports Microsoft
Cluster Server on
Windows Server 2003
• Supports Microsoft
Failover Clusters on
Windows Server 2008
and 2008 R2
File Share Witness with
RecoverPoint/CE
installed
CG1: Devices for
Cluster Group1
CG2: Devices for
Cluster Group2
• Supports synchronous
and asynchronous
replication
• Not needed for
Microsoft Exchange
Server 2010
Cluster nodes with Cluster Enabler 4.0 installed
© Copyright 2010 EMC Corporation. All rights reserved.
24
Local Protection Process—
Continuous Data Protection (CDP)
2a. Host splitter
1. Data is split and sent to the RecoverPoint
appliance in one of three ways
2b. Intelligent-fabric
splitter
3. Writes are acknowledged back
from the RecoverPoint appliance
4. The appliance writes data to the
journal volume, along with time
stamp and application-specific
bookmarks
2c.
CLARiiO
N splitter
/A
/B
/C
rA
Production volumes
rB
rC
Replica volumes
Journal volume
5. Write-order-consistent data is
distributed to the replica
volumes
© Copyright 2010 EMC Corporation. All rights reserved.
25
Remote Protection Process—
Continuous Remote Replication (CRR)
1. Data is split and
sent to the
RecoverPoint
appliance in
one of three
ways
2a. Host splitter
2b. Intelligent-fabric
splitter
3. Writes are
acknowledge
d back from
the
RecoverPoint
appliance
6. Data is
received,
uncompressed,
sequenced, and
verified
7. Data is written
to the journal
volume
2c.
CLARiiO
N splitter
/A
/B
Local site
/C
5. Data is sequenced,
checked, compressed,
and replicated to the
4. Appliance functions
remote appliances over
• Fibre Channel-IP conversion
IP or SAN
• Replication
• Data reduction and
compression
• Monitoring and management
rA
rB
rC
Remote site
Journal volume
8. Consistent data is
distributed to the
remote volumes
© Copyright 2010 EMC Corporation. All rights reserved.
26
Synchronous Remote Replication: Zero
RPO
• Replication only supported
over Fibre Channel network
• Dynamically switches
between synchronous and
asynchronous
Prod
Fibre
Channel
Target
Synchronous
Production
site
Out-of-region
site
• Waits for acknowledgement
from the remote site
• All synchronization will be
asynchronous
• Supports RecoverPoint and
RecoverPoint/SE
© Copyright 2010 EMC Corporation. All rights reserved.
27
Dynamic Switching Between
Synchronous and Asynchronous
Set in a policy for each consistency group
Synchronous
replication
Switch based
on latency
Switch based on
production throughput
Throttle
production
application
© Copyright 2010 EMC Corporation. All rights reserved.
28
Synchronous Replication
Consistency group members indicate synchronization state
© Copyright 2010 EMC Corporation. All rights reserved.
29
RecoverPoint
Integration with Server
Virtualization
© Copyright 2010 EMC Corporation. All rights reserved.
30
Virtualization Platform Support
VMware ESX
Hardware
Resource
Pool
• Supports Windows
Server 2008, 2008
R2, and Hyper-V,
Hyper-V R2
(RecoverPoint and
RecoverPoint/Cluster
Enabler 4.0)
• Supports VMs
running on ESX 3.x
and ESX 4.0, ESX
4.1, and Hyper-V,
Hyper-V R2
• Integrates with
vCenter Server for
discovery, SRM, and
automation of SRM
failback
© Copyright 2010 EMC Corporation. All rights reserved.
31
VMware Affinity
vCenter Site Recovery Manager (SRM)
PRODUCTION
RECOVERY
Virtual Machines
Virtual Machines
VMware Infrastructure
VMware Infrastructure
Servers
Servers
Heterogeneous Storage
Heterogeneous Storage
EMC RecoverPoint
• VMware SRM consistency groups
– Can only be monitored
– Require maintenance mode for snapshot
access, but failover cannot occur in this
mode
• SRM uses latest image during its test
or disaster failover
– A restriction on vCenter Site Recovery
Manager not RecoverPoint
– CDP replica and journal not available during
a real failover (only vCenter Site Recovery
Manager test mode)
• EMC developed adapter integrates
with VMware SRM and supports
VMware vCenter Server
– Supports ESX 3.5 and ESX 4.x
© Copyright 2010 EMC Corporation. All rights reserved.
32
Automated Failback After SRM Failover
• Integrates into vCenter Server
• Automates failback by:
– Powering down VMs at recovery
site
– Replicating LUNs from recovery to
protected site
– Reestablishing replication
between protection and recovery
sites
– Registering VMs
– Powering on VMs at protected site
• Provides real-time status
• Controlled in vCenter Server
• Available in the “Recent Tasks”
area of the vCenter Server
© Copyright 2010 EMC Corporation. All rights reserved.
33
VMware Integration with vCenter Server
• RecoverPoint queries vCenter Servers for list of
VMs
• Protection status is shown on RecoverPoint GUI
• Protection status is available from RecoverPoint
CLI
• RecoverPoint issues an alert for unprotected VMs
– Existing VM not protected by RecoverPoint
– Existing VMs storage moved (e.g., Storage vMotion)
– New VM created that is not protected by RecoverPoint
• Optional filtering (e.g., ignore checking protection
status) by ESX server, VM or LUN
© Copyright 2010 EMC Corporation. All rights reserved.
34
VMware Affinity
Sample display
© Copyright 2010 EMC Corporation. All rights reserved.
35
Continuous Replication
and Application
Recovery
© Copyright 2010 EMC Corporation. All rights reserved.
36
Snap and Clone Comparison
RECOVERPOINT
Production
Replica
RecoverPoint
Snaps/Clones
Multiple point-in-time images
Limited number of predefined
point-in-time images
All writes captured continually
Images taken every few hours
Journal contains point-in-time
images
Number determined by user,
disk space
Journal
TRADITIONAL SNAPS/CLONES
Production
RecoverPoint compared to array snaps/clones
Full
copy
Significant bandwidth
Limited bandwidth reduction
reduction for remote replication when replicated
Specify different RPO/RTO
and policies per application
RPO/RTO changes require
scripting per application
Consolidation saves storage
space
Storage requirements grow
depending on image type
Delta
copy
Original
production data
© Copyright 2010 EMC Corporation. All rights reserved.
37
Continuous Recovery Points
Daily backup
Daily recovery points—from tape or disk
Snapshots
Any point in time
More frequent disk-based recovery points
All recovery points
Significant point in time
Database
checkpoint
Pre-app
patch
Post-app Database
patch
checkpoint
Quarterly
close
Any userconfigurable event
Significant
points in time
RecoverPoint
Any point
in time
Snapshot
Daily backup
24 hours
Yesterday
© Copyright 2010 EMC Corporation. All rights reserved.
Midnight
Now
38
VSS Support of Microsoft Exchange
Server
Example
PRODUCTION SITE
DISASTER RECOVERY SITE
Exchange 2003 Cluster
Exchange 2003
Recovery Servers
4. VSS backup
complete—return
Exchange to
normal
operations
2. Every three hours
run KVSS to
place Exchange
into VSS backup
mode
1. Group EX1
(SGs + Logs)
continuously
replicated to
disaster recovery
site with a
15-minute
guaranteed RPO
© Copyright 2010 EMC Corporation. All rights reserved.
3. When Exchange is ready for VSS
backup, close latest
RecoverPoint image and tag as a
VSS image
WAN
5. Latest Image sent to
disaster recovery site
(VSS IMAGE3) + VSS
Bookmark
6. Latest image
distributed
IMAGE 24
--IMAGE 13
VSS IMAGE
2
IMAGE 12
--- --IMAGE 2
VSS IMAGE
1
IMAGE 1
39
Journaling for Application-Aware
Recovery
Journal includes replicated data plus metadata
• Time/date:
– Exact time (down to milliseconds) that
the image was stored
• Size:
– Size of image
• Image bookmark details:
– System-, EMC-, or user- generated
bookmarks
• Bookmark consolidation options
– System decides (blank)
– Survive a daily, weekly, or monthly
consolidation
– Never consolidate
© Copyright 2010 EMC Corporation. All rights reserved.
40
Recovery Consolidation
Journal recovery point timeline
Continuous recovery points
Two days
ago
Now
Daily recovery points
Weekly recovery points
Monthly
Weekly
Daily
Continuous
Monthly recovery points
RecoverPoint snapshot
consolidation…
 Provides longer recovery
windows with same storage
 Optimizes usage of journal
storage
 Helps eliminate concerns for
backup windows
© Copyright 2010 EMC Corporation. All rights reserved.
41
Recovery Process
Servers at recovery site
Recovery
request
SAN
RecoverPoint
appliance
Journal Target volumes vLUN
• Multiple use cases:
– Testing of production failover
– Starting new application development
– Surgically repairing data
– Fast production resynchronization
– Source for backup, data analysis,
disaster recovery, and
development/test
• Request recovery via GUI, CLI, or
through integrations with
NetWorker, Replication Manager,
SRM, or Cluster Enabler 4.0
• Image recovered as read/write
• Any change to image saved in
journal
© Copyright 2010 EMC Corporation. All rights reserved.
42
Overview on Data Consistency
• Applications and data are
interrelated (federated)
DB
• All data movement must be
stopped/ started at the same point
in time
Order Entry
CRM
DB
DB
SCM
• To restart applications, you must
have all the data—not parts of it
• Recovery requires dependent-write
consistency across all volumes and
systems
Systems share information…
RecoverPoint captures a consistent view
© Copyright 2010 EMC Corporation. All rights reserved.
43
Consistency Groups (CG)
• RecoverPoint ensures consistency
across production and targets
OE
CRM
CDP
SCM
CDP
E-mail
CG 1
CG 2
CRR
• Application recovery can be
independent of SLAs
CRR
• RecoverPoint enables independent
replication of various applications
CRR
CG 3
© Copyright 2010 EMC Corporation. All rights reserved.
CRR
• RecoverPoint utilizes consistency
groups to organize data
44
Group Sets protect Federated
Environments
1: Linux (Web OE)
Consistency
group
2: Windows (CRM)
Consistency
group
• Each tier may have different
service level agreements
• Enforces write-order across
tiers
• Used for functions such as
upgrades, backups, and data
mining
3: UNIX (SCM, Financials…)
Consistency
group
© Copyright 2010 EMC Corporation. All rights reserved.
45
Defining Group Sets
© Copyright 2010 EMC Corporation. All rights reserved.
46
Supporting Federated Environments
Customer X Customer Y
added
added
1
2
Order X1
received
3
Receivable X1
added
4
5
Order Y1
received
6
Receivable Y1
received
BUSINESS APPLICATION
Order
Entry:
Web OE
Customer
Records:
SQL
Supply
Chain:
Oracle
1 2 3
4 5 6
2 4
Mid-tier storage
© Copyright 2010 EMC Corporation. All rights reserved.
1
3
5
6
Journal
High-end storage
47
Systems Monitoring,
Management, and
Security
© Copyright 2010 EMC Corporation. All rights reserved.
48
Monitoring and Alerting for Limits
• Lets users quickly see their protection limits
• RecoverPoint GUI shows where the user’s current
configuration stands relative to the system limits
• RecoverPoint CLU lets the user receive a list of
items, their limits, and their current values
• Crossing thresholds triggers events and a call
home which improves supportability
© Copyright 2010 EMC Corporation. All rights reserved.
49
Monitoring and Alerting for Limits
Sample display
© Copyright 2010 EMC Corporation. All rights reserved.
50
Management
RecoverPoint Management
Element-Management Tools
Element
Tool
Servers
Server-management tools
SAN fabric
SAN-management tools
RecoverPoint
RecoverPoint CLI and GUI,
Replication Manager, SRM,
NetWorker, Cluster Enabler
Intelligent
switches
Brocade: Fabric Manager
Cisco: Fabric Manager
Arrays
Unisphere with RecoverPoint
3.3 SP1 or other arraymanagement tools
IP network
Network-management tools
© Copyright 2010 EMC Corporation. All rights reserved.
• Not a replacement for other
element tools
• Java-based interface
• Command line interface
• Script interface
• Internal notices sent via
SNMP, syslog, and e-mail
• Dial home capability
51
Management Application GUI
• Administration and
management
• Update configuration
• Monitoring and diagnostics
Administrator
Network
• Java-based, does not require
a dedicated management
server
• LDAP with access control
list-based authentication
• Simultaneous access across
enterprise
© Copyright 2010 EMC Corporation. All rights reserved.
52
Online Help for Management Application
• Accessed via “Help > Help
Contents” from the main
RecoverPoint GUI
• The contents are based on
existing RecoverPoint
documentation
• Content is downloaded to
the local computer
• Users can view, search,
bookmark, and print
sections
© Copyright 2010 EMC Corporation. All rights reserved.
53
Management Using Unisphere
Unisphere Management for RecoverPoint/SE
• Manage RecoverPoint/SE
from the same management
console used to provision
CLARiiON
• Requires FLARE 30
• Supported on
RecoverPoint/SE 3.3 SP1 or
later
• Requires installation via
Deployment Manager
© Copyright 2010 EMC Corporation. All rights reserved.
54
Virtual Provisioning Support
Available for CLARiiON CX4 and Symmetrix VMAX, DMX-4,
DMX-3
Reported
capacity
Allocated
SAN/WAN
Allocated
• RecoverPoint is thin aware
Common
storage
pool
– RecoverPoint recognizes thin LUNs
– Preserves allocation of thin LUNs—during
initial synchronization, resynchronization
– Allows mixing of thin and thick LUNs
Data
devices
Data
devices
 Replication between CLARiiON CX4 and nonCX4
 Replicate between different Engunity versions
Symmetrix VMAX,
DMX-4, DMX-3
– Virtual-to-virtual (thin-to-thin) replication
CLARiiON CX4
• RecoverPoint is licensed by LUN
capacity presented to the SAN
RecoverPoint understands allocated storage
© Copyright 2010 EMC Corporation. All rights reserved.
55
RecoverPoint Security
• Appliance runs as a hardened
configuration
• DMZ and firewall compliant
• Login and authentication enforced and
logged
• Event logging for audit purposes
• Multiple administrative roles
• LDAP authentication
• ACLs and role-based controls
• No customer root access
© Copyright 2010 EMC Corporation. All rights reserved.
56
Systems Integrity
• An RPA is not general
purpose, it:
Administrator
Secure
network
RPA
© Copyright 2010 EMC Corporation. All rights reserved.
– Runs RecoverPoint software
– Does not run customer
binaries
– Is supported as a single unit
• RecoverPoint appliance GUI
– Secure management path
– Administrator communicates
through GUI or CLI
– RecoverPoint appliances
exchange health and status
57
Administrative Security
• Supports local or LDAP/AD user
authentication
• Security role to add/edit/delete
users
• Role-based access controls
• Users can be limited to specific
consistency groups and/or by
responsibility
• User create and modify
performed in LDAP/AD; settings
imported by RecoverPoint
© Copyright 2010 EMC Corporation. All rights reserved.
58
Event Logging
• Warnings and error events show
up in top window with an icon
• Information, warning, and error
messages viewable through GUI
• All events logged by
RecoverPoint
• All events can also be filtered
and then sent via SMTP, e-mail,
or syslogs
© Copyright 2010 EMC Corporation. All rights reserved.
59
The Industry-Leading Network-Based
Replication Solution
RecoverPoint/SE
RecoverPoint
CLARiiON only
Easy to deploy
Simple to manage
Heterogeneous storage
Expanded protection capacity
Supports intelligent fabric
Availability
Failover
Advanced clustering
Number of RecoverPoint
appliances
2–8 per side
2–8 per side
Write splitting
Windows host,
CLARiiON CX4 or CX3
IBM AIX, Sun, Windows host,
CLARiiON CX4 and CX3, Brocade
and Cisco APIs
Storage
Single CLARiiON per
side
Multiple CLARiiON, Symmetrix, thirdparty per side
Journal compression
None
None, medium, high
Maximum replicated storage
150 TB
600 TB
© Copyright 2010 EMC Corporation. All rights reserved.
60
A Heterogeneous, Multi-Vendor,
Standards-Based Solution
Industry leaders as partners
Robust heterogeneous system
qualification
EMC
Oracle Solaris
IBM AIX
HP
Microsoft Windows
HP-UX
HDS
Red Hat/SUSE Linux
OpenVMS
IBM
VMware
Consult the “EMC Support Matrix” on EMC.com for
qualification and support details.
© Copyright 2010 EMC Corporation. All rights reserved.
61
Summary
• Lowest total cost of ownership
• Integrated CLARiiON support
• Intelligent SAN-switch
integration
• Any point-in-time recovery
• Highest availability, highest
performance
• Heterogeneous support for
interoperability
© Copyright 2010 EMC Corporation. All rights reserved.
62
THANK YOU
© Copyright 2010 EMC Corporation. All rights reserved.
63

similar documents