Next Generation Datacentres

1
(No Transcript)
2
Next Generation Datacentres

• Ulrich Homann, Microsoft Corp
• Ian Erridge, Microsoft Ltd.

3
Agenda

• State of the Industry
• The Dynamic Services Platform (DSP) solution
• Benefits Reasons to choose
• Questions?

4
After Lunch Quiz
5
State of the Industry Software Services
Web 2.0
Service
SaaS
SOA
6
State of the Industry Most Datacentres
Staffing
Over-provisioning
Power
Space constraints
Under-utilized resources
Regulatory Compliance
Lack of agility
Demand for distributed, always available services
7
Datacentre Today Management Of Servers

• Realities
• Manual processes
• Service domains work independently of each other

• Results
• High admin to server ratio
• Extended time to production/ market
• Inflexible processes and vendors
• Reactive management
• Lack of insight into application demographics
• High number of change management tickets
• Decreased service availability

Service
Server
8
We Hear You Customers Are Saying Our Needs
Are
9
Datacentre Tomorrow Moving From Servers To
Services

• Desired results
• Optimized admin to server ratio
• Decreased time to production/ market
• Automated processes
• Proactive management
• Greater insight into application demographics
• Streamlined change management process
• Higher service availability

• Focus
• Automated processes
• Greater coordination between service domains

Service
10
Dynamic Services Platform (DSP) solution

• Microsoft Services view of a service-oriented
  datacentre operating model, providing an
  end-to-end view and approach to delivering
  services out of a datacentre

11
DSP Solution Approach
Encoded in software models (SML)

• Knowledge of the service
• Architectural intent
• Operational environment
• IT policies
• Resource needs

Core Concepts(Hardware, Software,relationships)

• That can be created, modified and operated on ...
• Across the IT Lifecycle
• Develop, Operate, Analyze/Act

12
Solution Approach
13
Why Services?
14
Resource Consumption External and Internal
15
DSP Solution Overview
16
DSP solution overview
The Dynamic Services Platform (DSP) is a service
focused data center automation solution that
addresses the end-to-end service lifecycle,
dramatically reducing cost of ownership of data
center operations
Service Provisioning
Service Management

• Automation of routine tasks
• Patch management
• Managed remediation
• Migration
• In-service routine maintenance
• Versioning and rollback
• Auditing and reporting
• Business Intelligence

• Model
• Discovery
• Templates
• Template validation tool
• Provisioning lifecycle
• Self-deployment
• Versioning and rollback

17
DSP Solution Architecture
Solution Components
DSP DC-Scale SharePoint
DSP DC-Scale SQL Server
Modeling Tools
Process Templates
DSP Client UI Tool
IT Service Model
DSP Platform Components
Partner/ Custom Platform Components
IT Service Manager for Distributed Applications
IT Service Manager Pattern
Process Execution Environment
Physical / virtual server environment housing
multiple services and applications
18
DSP Design-Time Run-Time
1
Service Designer models service
DSP models are loaded from repository and
executed by DSP runtime
2
DSP infrastructure
built using
ITService Manager
Workflows
3
Service workflows execute PS scripts
4
ITService Management application identifies
workflows and other artifacts
PowerShell Scripts
19
Expanding the Service Model
Data Contract (SML)
IT Service Model
Policy / Rules (XOML)
.Net Java Assemblies
Process / Workflow
WebServices WS-Man
PowerShell VB CMD SSH
XOML
20
The DSP Data Model
Data Contract (SML)
ltITServicegt
ltSitegt
Implementation
Functionality
ltCapabilitiesgt
ltClustergt
1..N
ltRoleServicegt
ltServerRolegt
ltWorkLoadgt
1..N
1..1 instance
21
Service Configuration Developer Experience
Integrated WorkFlow PowerShell Development
22
Expanding the Modeling Tools
23
Expanding the Templates
24
Projected Load Profile
25
Load by Time of Day
26
Going Green
27
Guided Provisioning

• Petes SharePoint order (representing max
  growth)
• 50,000 users
• 20,000 team sites
• 150MB/site
• Responses per second 100

• Monitoring counters in the operational
  configuration and monitoring environment (SC OM
  2007) trigger growth (or shrink) provisioning
  once the specific capacity driver hits 80 of
  specified value
• Growth based upon RPS (growth type A) initial
  size 99 RPS counter is set to 80 RPS
• Growth based upon content db size (growth type
  B) initial size 0.8 TB counter is set to 0.7
  TB

28
A Responsible Dynamic Topology
29
Implications of Green Computing

• The goal in most green initiatives has to be
  focused on using only the minimally required
  resources for the given workloads.
• Ideally, systems will be deployed with the
  minimally required resources and allow for
  dynamic growing and shrinking of required
  resources based upon load heuristics.
• Consolidate workloads onto fewer resources by
  analysis of load patterns and hotspots
• Permanently and Temporarily
• Shut down non-required resources
• What is the Architecture for a Green Application?

30
Questions
31
(No Transcript)