HNAS Replication Overview

Report
HITACHI NAS
REPLICATION
OVERVIEW
NOVEMBER 2013
1
COMMON IT BUSINESS CHALLENGES
 Protecting information
given exploding data growth
 Delivering a powerful and
highly available infrastructure
 Balancing desired data
protection RPO and RTO with
costs and budgets
 Understanding and preparing
for all types of disasters and
outages
 Testing and assuring disaster
recovery plans
2
SEVERE RAMIFICATIONS OF DATA LOSS
93% of companies that lost the use of their data center for 10 days or more filed for
bankruptcy within one year of the disaster
(Source: National Archives & Records Administration, Washington D.C.)
Of the companies experiencing disasters, 43 percent never reopen, and 29 percent
close within two years
(Source: McGladrey & Pullen, LLP).
The average cost of downtime as a function of labor (for large organizations) is
$1,010,536 per hour
(Source: Meta Group Study).
A service interruption could cost a company (Bank) between $60,000 and $250,000
a minute
(Source: U.S. Banker).
Data unavailability results in lost revenue and possibly business failure
3
REPLICATION IS THE ANSWER
HITACHI NAS REPLICATION DELIVERS MULTIPLE BENEFITS
 High Availability
‒ Fault-tolerant, failure-resistant streamlined infrastructure
with affordable cost foundation
 Continuous Operations
‒ Non-disruptive backups and system maintenance
coupled with continuous availability of applications
 Disaster Recovery
‒ Protection against unplanned outages such as disasters
through reliable, predictable recovery
4
HITACHI NAS REPLICATION CHOICES
HNAS HAS THREE REPLICATION MECHANISMS AVAILABLE
 File Level Replication
‒ NDMP based data copy of files
‒ Walks directory tree to build delta/incremental
 Block Level Replication
‒ Leverages underlying storage replication technology
‒ Enables High Availability and advanced replication scenarios
 Object Level Replication
‒ Only replicates objects changed since last checkpoint
‒ Results in faster, more efficient replication
5
OBJECT REPLICATION IS USUALLY BEST
Ideal for DR and data migration
Ideal for
fine
granularity
when backup windows are smaller
Up to 2.8 times faster than file
replication on full file replication
File Replication: Granular
Block Replication: Powerful
Ideal for high
availability
configurations
Object Replication:
Fast and Efficient
Data
Data
Object
Object
Data
Data
Object
Data Object
Data
Object
Object
Only tracking
incremental
changes
Eliminates traversing directories
and files for small changes
No wasted empty containers sent
6
OBJECT LEVEL REPLICATION OVERVIEW
HIGH PERFORMANCE HARDWARE ACCELERATED REPLICATION
Primary File Server
HNAS Server
DR File Server
Network
HNAS Server
Direction of replication
File system A
Replication Source
File system B
Replication Target
 Snapshots on source have equivalent snapshot on target
 Target file system is not mountable unless promoted
 Multiple node clusters can replicate to a single node
cluster or one cluster can replicate to multiple clusters
7
HIGH PERFORMANCE OBJECT REPLICATION
8
Test Case
File Level
Replication
Object Level
Replication
Performance
Improvement
Full replication of
40,000 large files
(100 MiB)
4 hrs, 35 m, 3 s
242.38 MiB/s
1 hr, 39 m, 33 s
669.70 MiB/s
2.8 times
faster
Full replication of
a large number
(8,000,000) of
small files (4 KiB)
38 m, 20 s
3,478 files/s
20 m, 13 s
6,595 objects/s
1.9 times
faster
Delta replication
(16%) of a large
number (8 million)
small files (4 KiB)
59 m, 2 s
361 files/s
2 m, 16 s
9,412 objects/s
26 times
faster!
BLOCK REPLICATION FOR HIGH AVAILABILITY
ACHIEVE ZERO RTO BY LAYERING HA AND SYNCHRONOUS REPLICATION
 Concept
HOST
HOST
Application
Application
Multi path Software
Multi path Software
‒ Continuous availability solution
in a local or remote site
 Configuration
Host paths
VSP/HUS
VOL pair
PVOL
‒ Mirrored internal and external data
and controller
VSP/HUS
SVOL
TrueCopy paths
UVM
UVM
Quorum
Any
External Storage
 Benefit
‒ Eliminate business impact associated
with applications downtime
‒ Applications can run continuously
despite unplanned outage of storage
(and/or controller)
‒ Simplified and non-disruptive
migrations to next-generation
controllers
9
BLOCK REPLICATION MANAGEMENT
10
FILE REPLICATION FOR FINE GRANULARITY
REPLICATE ONLY THE FILES OR DIRECTORIES YOU REALLY NEED
 Concept
‒ Specifically chose which files and/or
directories you wish to replicate
 Benefits
‒ Eliminates time and resources wasted
on replicating unneeded files or
directories
‒ Allows informed decisions on what to
replicate
‒ Ability to exclude temporary files or
parts of a file system that hold no
business value
‒ Substantially reduces the amount of
data sent from the source machine
‒ Decreases the storage requirements of
the destination system
11
CREDIT UNION AUSTRALIA SUCCESS STORY
“To become the 5th pillar of the Australian banking
system and compete with bigger players in the
industry, our organization is going through a major
transformation. The platform upgrade was businesschanging.”
David Gee, Chief Information Officer, Credit Union Australia
CHALLENGES
 Data center required an
upgrade to enable
sustainable growth
 Need for big-bank
platform quality and
power on a scalable
and flexible model, to
effectively compete with
the “big 4” banks
12
HDS SOLUTION
 Hitachi Virtual Storage
Platform
 Hitachi Unified Storage
 Hitachi Content
Platform
 Hitachi NAS Platform
RESULTS
 Ability to replicate all its
data securely and
safeguard against disaster
situations
 Flexibility of replication
service with responsiveness
that could maintain
availability at 99.98%
 Ability to scale and react to
business demand
PRECISION DISCOVERY SUCCESS STORY
“A big reason we’ve stayed with Hitachi all these years
is its people: The willingness and expertise from
Hitachi engineers, sales, field and support keep us
coming back. They have earned our complete trust
and loyalty.”
Howard Holton, Director of IT Precision Discovery
CHALLENGES
13
HDS SOLUTION
RESULTS
 To provide high caliber
service while offering
innovative technology &
expert consulting
 Hitachi NAS Platform
3090 & 3080
 Ability to manage 2PB of
storage
 Hitachi Unified Storage
 Ability to alleviate the
typical challenges of ediscovery
 Hitachi Compute
Blades 320, 500 & 2000
 Reclaim storage in the
environment, & reduced
operational expenses
 Hitachi Content
Platform
 Ability to handle data
reduction, deduplication,
replication & compression
HITACHI REPLICATION HIGHLY RATED
9 of 10
Rate HDS replication software
as effective or very effective in
meeting requirements*
85%
Users indicate positive
ease-of-use experience
Excellence in
Customer Service
Awards from:
* Source: TechValidate HDS Replication Use-case Survey October 2013
14
HNAS REPLICATION VALUE PROPOSITION
 Hitachi NAS provides three powerful methods of
replication to meet different data protection, high
availability and/or disaster recovery needs
 Benefits:
‒ Reduces data loss risks via proven
enterprise-class replication technologies
‒ Delivers SLAs that are better aligned with
business objectives
‒ Cost savings through:
‒ Lower network bandwidth requirements
‒ Reduced energy, cooling and space
‒ Simpler management and maintenance
‒ Increased uptime of mission critical systems
15
QUESTIONS AND
DISCUSSION
16
THANK YOU
17

similar documents