performance and scalability. performance and scalability challenges optimizing performancescaling...
TRANSCRIPT
Performance and Scalability
Performance and Scalability Challenges
Optimizing Performance
Scaling Up
Scaling Out
Performance and Scalability SQL Server 2008
Real-world, predictable performance
Scale up with today’s hardware
Scale out for the enterprise
TPC benchmarks
Workload prioritization
Tuning and optimization
tools
Enterprise health
monitoring
Improved BI performance Multi-
instance architecture
6464-bit
technologies
NUMANUMA
Support
+Hot-add hardware
Scalable shared
databases
Distributed partitioned views
Peer-to-peer replication
Query notifications
Service Broker
Data dependent
routing
Scalable shared
databases for Analysis Services
Optimizing Performance
Scaling Up
Scaling Out
Optimizing Performance
Optimizing Performance
Relational Database Performance
Measurable, real-world
performance
• TPC-E• TPC-H• SAP-SD
Performance
optimization tools• SQL Server
Profiler• Database
Engine Tuning Advisor
High performance query processing engine
• Cost-based optimization
• Read-ahead scans
• Partitioning
Admin Workload
Backup
Admin Tasks
Report WorkloadOLTP Workload
OLTP Activity
Executive
ReportsAd-hoc Reports
High
SQL Server
Min Memory 10%Max Memory 20%
Max CPU 20%
Admin Pool
Max CPU 90%
Application Pool
Ability to differentiate workloads such as app_name, login, and so onPer-request limits:
Max memory %Max CPU timeGrant timeoutMax requests
Resource monitoring
Relational Database PerformanceResource Governor
• Data providers• SQL Trace• Performance Counters• Transact-SQL
Low overhead performance data collection
Relational Database PerformancePerformance Studio
Data Warehouse and Analysis Performance
Data Warehouse Performance
• Star join query optimizations
• Improved parallelism• Partition aligned indexed
views• Data compression• MERGE• GROUPING SETS
Analysis Services Performance
• Block computations• Write-back to MOLAP
Reporting Performance
On-
demand processing
Effective caching and
memory management
Re-engineered rendering architecture
Reporting Services
ETL Performance
High performance lookup operations
Pipeline performance
Scaling Up
Hardware Support
Concurrency
Hardware Support
64-bit
Multi-core, multi-processor
Large-scale memoryNUMA hardware
Hot-Add hardware• Memory• Processor
Concurrency
Transaction isolation
•Read uncommitted•Read committed•Repeatable read•Serializable•Read committed (with row versioning)•Snapshot
Lock escalation granularity
•Row•Page•Partition•Table
Distributed Partitioned Views
Peer-to-Peer Replication
Query Notifications
Scalable Shared Databases
Scalable Shared Databases for Analysis Services
Scaling Out
Scalable Shared Databases
Read-only database in SANMounted by multiple reporting serversApplications access a consistent copy from any server
Distributed Partitioned Views
Data is partitioned horizontally across multiple serversTransact-SQL view retrieves all data with a UNION ALL clauseRequests can be directed by using data dependent routing
Peer-to-Peer Replication
Data is replicated to local serversLocal modifications are propagated throughout the enterprise
Query Notifications
Application is notified when queried data changesOptimal cache refresh behavior
Scalable Shared Databases for Analysis Services
Centralized, read-only Analysis Services database shared by multiple instancesClient applications connect to a single virtual IP address
© 2009 Microsoft Corporation. All rights reserved.This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.