Mission-Critical Operations

“We have multiple tables that exceed 1 billion rows. After big games we sometimes see loads of 1,500 customers per second querying the system and SQL Server supports the demand nicely” – bwin

Get measurable, real-world performance improvement.

• Take advantage of demonstrated high-performance capabilities with Transaction Processing Performance Council’s TPC-C and TPC-H benchmarks for OLTP and data-warehousing workloads.

• Gain support from the first database vendor to publish results for the newer TPC-E benchmark, which more accurately represents the kinds of OLTP workloads common in modern organizations.

Improve performance with the query-processing engine

• Generate optimal query execution plans based on dynamically maintained statistics about table, indexes, and data volume.

• Reduce the amount of physical disk access with dynamic caching.

• Use read-ahead scans to anticipate the data pages that are required for a given execution plan and preemptively read them into the cache.

• Take advantage of multiprocessor servers by using query parallelism.

Take control of resource allocation with Resource Governor

• Provide predictable performance across workloads by defining resource limits and priorities for workloads.

• Manage multiple-workload environments with workload prioritization.

• Minimize the effect of runaway queries on other workloads.

• Ensure that mission-critical workloads have the resources they need.

Troubleshoot, tune, and monitor with Performance Studio

• Integrate the collection, analysis, troubleshooting, and persistence of SQL Server diagnostics.

• Use the low-overhead data collector to gather performance data.

• Centralize storage of performance data from across the enterprise in a management data warehouse.

• Analyze performance data in a Performance Dashboard.

“With our 64-bit SQL Server 2008 running on Itanium 2 processors, and the ability to add all the memory we could need, scalability isn’t an issue, even for our rapid data growth.” – Premier Bancard

Gain support for the highest level of hardware scale

• Support more than 8 processors.

• Handle up to 256 logical processors.

Get support for 64-bit technologies to best use high-end servers.

• Provide greater scalability for systems with support for nonuniform memory access (NUMA) hardware.

• Take advantage of NUMA-based computers without application configuration changes.

• Gain support for both hardware NUMA and software NUMA.

Get support for up to 50 instances on one server

• Optimize hardware usage by supporting multiple Database Engine and Analysis Services instances on a single Windows Server–based machine.

“The near-linear scaling across 256 Intel Xeon 5400 processor with a total of 1,024 cores provides a level of performance that can meet the needs of the largest global financial institutions” – SunGard

Benefit from scalable shared databases

• Create a dedicated read-only reporting database by using the scalable shared-database feature.

• Provide enhanced performance by mounting these read-only databases on multiple reporting servers.

• Offer all users an identical picture of the data, no matter which reporting server is used.

Take advantage of peer-to-peer replication

• Scale out heavily accessed databases while maintaining consistency.

• Enable users to update multiple copies of the database while maintaining consistency.

• Reduce implementation and management time for peer-to-peer replication with the new visual designer.

Gain efficiency with query notifications

• Build applications to efficiently use application cache by taking advantage of the ability to notify middle-tier applications about outdated cache.

Build highly scalable, service-oriented solutions with Service Broker

• Create applications that use the Database Engine components to communicate between databases.

• Distribute data workloads across databases without developing communication and messaging internals.

• Handle the routing of partitioned data to multiple databases, services, or instances.

• Gain support for load balancing, forwarding, and gateways to increase reliability and performance to a distributed database.