Transcript Scalability, Caching and Elasticity

Create as many
roles as
you need
‘knobs’
to adjust
Consolidation
of Roles
provides
more
redundancy
forscale
same
cost
Clean Up Role
Queue Drive
CacheRole
Build Role
Web Site Role’
Web Driven Role WCF Role
Loss
instance
results
50%
Loss
of of
anan
instance
results
in in
just
25%
capacity loss in web site.
Size
CPU Cores
Network
RAM
Local
Storage
Cost
Small
1
Shared
1.7GB
250GB
1x
Medium
2
Shared
3.5GB
500GB
2x
Large
4
Shared
7GB
1000GB
4x
Extra large
8
Dedicated
15GB
2000GB
8x
v1.0
v2.0
?snapshot=<DT1>
?snapshot=<DT2>
Dedicated compute cache roles must serve at least 120,000 cache requests
per hour to be cheaper than Windows Azure storage
Outside USA and Europe: use CDN for caching due to much lower
bandwidth costs
Inactivity
Period
Usage
Average
Compute
Compute
Elastic Cloud Workflow Patterns
Time
Time
Compute
Compute
Time
Average Usage
Average Usage
Average Usage
Time
http://tinyurl.com/perfmon-overhead