Professional Documents
Culture Documents
Disaster
Recovery
3
Data Center
Virtualization
Network Consolidation
Compute Internet Optimization
Evolution Computing Data Center
Client/ Networking
Server 1.Consolidation
Mainframes 2.Integration
Content 3.Virtualization
Networking 4.High Availability
Thin
Client:
TCP/IP HTTP
Network
Terminal Evolution
Metro Network
DWDM/SONET/Ethernet
FC
Tape SAN Secondary Data Center 4
What Is Distributed Data
Center?
Data Replication
FC FC
5
Front-----End IP Access
Layer
“Content Routing”
Site Selection
App A App B App A App C
FC FC
9
Application and Database
Layer
“Content Switching”
App A App B Load Balancing App A App C
“Server Clustering”
High Availability
FC FC
:
Backend SAN Extension
App A App B
“Storage” and “Optical” App A App C
Data Replication
and Transporting
FC FC
Disaster Recovery
• Recovery of data and resumption of service
—Ensuring
• business can recover and continue after
failure or disaster
• Ability of a business to adapt, change and
continue when confronted with various
outside impacts Mitigating the impact of a
disaster
3
Disaster Recovery
What It Means for Business
-
4
Time
Recovery Point Recovery Time
time t0 Time t1 Time t2
:
Network Failures
Service Internet Service
ISP failure Provider A Provider B
Site Failures
• Partial site failure Service
Provider A
Internet Service
Provider B
-Application maintenance
-Application migration
-Application scheduled
-DR exercise
• Complete site failure
-Disaster
4
Warm Standby
• A data center that is equipped with
hardware and communications
interfaces capable of providing
backup operating support
• Latest backups from the production data
center must be delivered
• Network access needs to be activated
• Application needs to be manually started
Disaster Recovery—
Active/Standby
IP/Optical Network
FC FC
Secondary
Primary Data Data Center
Center (Warm Standby)
Hot Standby
A data center that is environmentally ready and
has sufficient hardware, software to provide data
processing service with little down time
• Hot backup offers disaster recovery, with little or
no human intervention
• Application data is replicated from the primary
site
• A hot backup site provides better RTO/RPO than
warm standby but cost more to implement
• Business continuance
5
Disaster Recovery—
Active/Standby
IP/Optical Network
FC FC
9
Active/Active DR Design Multiple
Tiers of Application
Presentation Tier
Application Tier
Storage Tier
:
Active/Active Data Centers
Internal
Service
Internet Service
Network
Provider A Provider B Internal
Network
Active/Active Web
Hosting
Active/Active
Application
Processing
Active/Standby
Database Processing
or
Active/Active
for Different
Application 2
Site Selection Mechanisms
• Site selection mechanisms depend on the
technology or mix of technologies adopted
for request routing:
1. HTTP redirect
2. DNS-based
3. L3 Routing with Route Health Injection (RHI)
• Health of servers and/or applications needs
to be taken into account
• Optionally, other metrics (like load) can be
measured and utilized for a better selection
HTTP Redirection—Traffic Flow
http://www.cisco.com/
http://www1.cisco.com/
.1
TP/1 .com
T o
T/H .cisc ed
G E o v om
Kee
w M
1. t: ww 302 cisco
.c
s .1
Ho /1 w2
.
paliv
T TP w
w
2. H tion:
e
a
Loc
s
3. GET/H
TTP/1.1
Host: ww
w2.cisco
.com
HTTP/1.1
200 O K
http://www2.cisco.com/
DNS----Based Site Selection
—Traffic Flow Authoritative Name
Root Name Server for/
Server for .com
2
3 4 Authoritative
Name Server
DNS Proxy 5 cisco.com
1 6
10 7
8
9 Authoritative
Client
Name Server
http://www.cisco.com/ www.cisco.com
s
UDP:53 alive
s
ep
Ke
live
TCP:80
ep a
Ke
Data Center 1 Data Center 2
Route Health Injection
Implementation
Client A Router 11 Client B
Router 13
Router 10
2
Site Selection Summary
25
Cluster Overview
Load Balancing Cluster : multiple
copies of the same application
against the same data set,
usually read only Web Servers
High Availability Cluster :
multiple copies of application
that requires access to a
Application Servers
common data depository, usually
read and write
Clustering provides benefits for
availability, reliability, scalability, Database Servers
and manageability
2:
High Availability Cluster Design
Public Network : Client
/Application requests
APP
Cluster
Software
Cluster
Enabler
Private Network : OS
Interconnection between
nodes
3
HA Cluster Application View
Active/standby
Standby takes over when active fails
Two-node or multi-node
Active/active
Database requests load balanced all nodes
Lock mechanism ensures data integrity
Shared everything Node1 Node2
3
Geo--Clusters
Considerations
Geo-Cluster: Cluster That Span Multiple
Data Centers
WAN
Local Remote
Datacenter Datacenter
Node1 Node2
Disk Replication
Challenges:
Synchronous or Asynchronous
2 x RTT
Split brain
L2 heart-
beats
Storage 3
3
2
HA Cluster Challenges :
Split-Brain
Data Corruption
3
3
Layer 2 Heartbeats
WAN
Extended L2 Network : Local Remote
Datacenter Datacenter
L2 adjacency required
for node’s heartbeat. Public Layer 2 Network
Extending VLAN across Node1
Private Layer 2 Network
Node2
site is hazardous
Resolution : L3
Capability for Cluster Disk Replication
Synchronous or Asynchronous
Heartbeat. EoMPLS to
carry L2 hearbits
across DR sites.
3
4
Node1 Node2
Active Standby
RW WD
RW WD
Storage for Applications
• Presentation tier
Unrelated small data files commonly stored on
internal disks
Manual distribution
• Application processing tier
Transitional, unrelated data
Small files residing on file systems
May use RAID to spread data over multiple disks
• Storage tier
Large, permanent data files or raw data
Large batch updates, most likely real time
Log and data on separate volumes 35
Replication: Modes of
Operation
Synchronous
All data written to local and remote arrays before I/O
is complete and acknowledged to host
Speed of Light = 3 x 108m/s (Vacuum) ≈ 3.3Hs/km
Speed through Fiber ≈ ⅔ c ≈ 5Hs/km
2 RTT per write I/O = 20Hs/km
Asynchronous
Write acknowledged and I/O is complete after write to
local array; changes (writes) are replicated to remote
array asynchronously
39
Synchronous vs.
Asynchronous Trade-- Off
Enterprises Must Evaluate the Trade-Offs
Synchronous Asynchronous
Impact to Application No Application Performance
Performance Impact
Distance Limited (Are Unlimited Distance (Second
Both Sites Within the Site Outside Threat Radius)
Same Threat Radius)
Exposure to Possible Data
No Data Loss Loss
3:
Data Replication with DB
Example
• DB name
Control
• Creation date
Files
• Backup • Control files identify other files
performed
making up the database and
• Redo log time
period records content and state of the
• Datafile state db
• Datafile is only updated
Identify periodically
• Redo logs record db changes
resulting from transactions
Used to play back changes that
may not have been written to
Record datafile when failure occurred
Redo Log
Datafiles Changes To
Files
Typically archived as they fill to
• Table spaces M Database changes local and DR site destinations
• indexes
• Data dictionary
4
Data Replication with DB
Example (Cont.)
Time
SAN
Database Extension Database
Copy at
Copy at Transport Time t0
Time t0
Point in
Replicated/Copied
Time
Copy
Taken Archive Logs Archive Logs
When DB Replicated/Copied
Quiescent
nightly)
4
2
Data Center Interconnection
Options
Internet
Stateful Stateful Internet
Firewalls Firewalls
Content Content
Caching Caching
High SONET/SDH Server High
Server
Density Load Density
Load
Multilayer Balancing Balancing Multilayer
LAN LAN
Switch Intrusion Intrusion
Detection Switch
Detection
Front2End
Front2End Application
Application Servers
Servers
DWDM/
CWDM
Back2End
Back2End Application
Application Servers
Servers
High High
Density Density
Multilayer Multilayer
SAN SAN
Director Enterprise2Class
IP/Metro E Storage Arrays
Director
Enterprise2Class
Storage Arrays
4
3
Limited by BB_Credits
DWDM Sync (2Gbps Lambda)
Async
SONET/SDH Sync (1Gbps+ Subrate)
MDS9000 FCIP Sync (Metro Eth) Async (1Gbps+)
Cisco Data Center Vision
Server Enterprise AUTOMATION
Data Storage
Fabric Applications
Network Network Dynamic provisioning and
Network autonomic Information Lifecyle
Management (ILM) to enable
business agility
Business Policies
LAN HPC VIRTUALIZATION On2Demand
WAN SAN Service Oriented
Cluster Management of resources
MAN GRID independent of underlying
physical infrastructure to
Intelligent increase utilization, efficiency
Information and flexibility Compute
Network
CONSOLIDATION Network
Centralization and
standardization to lower costs,
improve efficiency and uptime Storage
Compute Network Storage
4
Today’s Data Centers
Require an Architectural Approach to…
4:
Embedded Intelligent
Virtual Fabrics (VSANs)
MDS 9000 Storage Services
Family Storage Virtualization
Grid/Utility Computing
Firewall Services Catalyst 6500 TOPSPIN
Family FAMILY Low Latency RDMA
Services
Intrusion Detection
Clustering
SERVER
ENTERPRISE FABRIC
WIN GRID SWITCHING
NAS UNIX