Mission Critical Performance and Scale with SQL Server and Windows Server

Technical White Paper

Published: September 2013 (updated)

Applies to: Microsoft SQL Server 2014 and SQL Server 2012

Summary: As the volume and complexity of data continue to increase, organizations require a new
approach to their mission-critical capabilities. In this white paper, we examine how capabilities built into
Microsoft SQL Server—including enterprise-grade security, availability, and performance, along with
support for a full range of complex data types—define a “new mission critical” that answers what kinds of
capabilities organizations need and expect to compete in a dynamic global landscape. We also compare
the cost impact of solutions that offer mission-critical functionality that is built into the core database with
solutions that offer separate features organizations can add at additional cost.
Copyright
The information contained in this document represents the current view of Microsoft Corporation on the
issues discussed as of the date of publication. Because Microsoft must respond to changing market
conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot
guarantee the accuracy of any information presented after the date of publication.

This white paper is for informational purposes only. MICROSOFT MAKES NO WARRANTIES,
EXPRESS, IMPLIED, OR STATUTORY, AS TO THE INFORMATION IN THIS DOCUMENT.

Complying with all applicable copyright laws is the responsibility of the user. Without limiting the rights
under copyright, no part of this document may be reproduced, stored in, or introduced into a retrieval
system, or transmitted in any form or by any means (electronic, mechanical, photocopying, recording, or
otherwise), or for any purpose, without the express written permission of Microsoft Corporation.

Microsoft may have patents, patent applications, trademarks, copyrights, or other intellectual property
rights covering subject matter in this document. Except as expressly provided in any written license
agreement from Microsoft, the furnishing of this document does not give you any license to these patents,
trademarks, copyrights, or other intellectual property.

© 2013 Microsoft Corporation. All rights reserved.

Microsoft, Active Directory, Azure, Excel, SharePoint, SQL Server, Windows, and Windows Server, are
trademarks of the Microsoft group of companies.

All other trademarks are property of their respective owners.

Page 2
Contents
The new mission critical................................................................................................... 5
Organizational requirements............................................................................................ 5
SQL Server.......................................................................................................................6
Data availability........................................................................................................................................ 6
High availability of mission-critical systems: SQL Server AlwaysOn..................................................................6
Online database operations.....................................................................................................................................7
Predictable, efficient, and flexible data backups....................................................................................................7
Better together with Windows Server....................................................................................................................8
Performance and scale............................................................................................................................ 9
In-memory online transaction processing (OLTP).................................................................................................9
In-memory data warehousing (DW): ColumnStore Index...................................................................................10
Buffer pool extension............................................................................................................................................11
Query-processing enhancements...........................................................................................................................11
Private cloud..........................................................................................................................................................11
Tier-1 partitioning: Scale to 15,000 partitions......................................................................................................12
Scalable real-world application testing: Distributed Replay................................................................................12
Reduced database size and increased performance: Data and backup compression............................................12
Proactive troubleshooting and diagnostics: Performance Data Collector and Management Studio....................12
Better together with Windows Server...................................................................................................................13
Organizational security and compliance................................................................................................. 14
Secure by default: Lowering vulnerability...........................................................................................................14
Built-in tools for enabling compliance: SQL Server audit enhancements............................................................14
Restricted access to data at the row level: Label security.....................................................................................14
Controlled access to data in motion: Advanced security......................................................................................15
Controlled access to data for administrators: User-Defined Server Roles...........................................................15
Backup Encryption................................................................................................................................................15
Controlled access to data across business intelligence tools: Microsoft SharePoint and Microsoft Active
Directory
..............................................................................................................................................................................
15
Virtually any data, built-in....................................................................................................................... 15
Embracing unstructured data through support for complex data types................................................................15
High availability for unstructured data.................................................................................................................16
Seamless connection and analysis of Big Data through Hadoop connectors.......................................................16

Page 3
 Seamless connection to a variety of platforms through greater interoperability..................................................16

Comparing the cost of mission-critical capabilities.........................................................17

Expecting the fundamentals................................................................................................................... 17
Understanding options........................................................................................................................... 17

Conclusion......................................................................................................................20

Page 4
The new mission critical
Data is growing everywhere, and the amplification of data affects every imaginable device, application,
and process as our world rapidly evolves into digital and virtual transactions and experiences. It is no
longer the case that organizations serve customers during standard business hours in single time zone or
geography. Today, services are continually available to customers through a range of operational
measures, from Internet presence at the very minimum to tracking complex operations globally for the
highest efficiencies and customer satisfaction. The notion of a maintenance window is no longer an
opportunity for IT to maintain and upgrade systems; these necessary activities are now problems.
Consumers, whether at work or at home, simply expect continuous service and access to information and
experiences on their terms. Mission-critical response is no longer reserved for the few who can afford to
pay astronomical costs to purchase and maintain Tier-1 systems.

Additionally, data amplification extends beyond traditional data types. According to Gartner, the total
worldwide volume of data is increasing 59 percent per year. Furthermore, Gartner estimates that 70–85
percent of data is unstructured.1 The dramatic shift from structured to unstructured and complex data
types requires organizations to embrace back-end solutions that support complex data types and
nontraditional data sources (such as Big Data) with the same level of support for mission-critical
capabilities.

IT organizations currently face the need to balance the impact of data amplification and the requirement
for global expansion with conservative budgets and ever-tightening compliance policies. They need to
accomplish this feat at higher levels of expected uptime and adherence with regulations that are
increasingly strict and diverse across geographies. More than ever, organizations need mission-critical
operations that are easy to deploy and are balanced with faster time-to-solution.

Organizational requirements
Our digital age demands a level of response to the demand for mission-critical capabilities that is no
longer reserved for the few who can afford costly Tier-1 systems. Organizations require mission-critical
capabilities across the following functions:
 Data availability: Solutions that are highly available across the globe and that fail over in
seconds, deliver reliable backups, and are easy to configure, maintain, and monitor—at a low total
cost of ownership (TCO).
 Performance and scale: Leading and predictable performance across server activity,
including complex queries, data integration, and analysis, and backed by scalable systems to
support increasing data.
 Security: Clean solutions that include inherently secure database software for reduced risk
combined with built-in, easy-to-use tools and controlled access to data that help organizations meet
strict compliance policies.
 Any data, built in: Support for growing volumes of complex data types and acceptance
of nontraditional data sources—with simple support for varied platforms and
heterogeneous environments.

1
Source: Gartner Symposium Presentation, “Information Management Goes 'Extreme'”: The Biggest Challenges for 21st-Century
CIOs, Mark Beyer, October 2011.

Page 5
SQL Server
SQL Server sets a new standard for mission-critical platforms by offering organizations the uptime and
performance they need at a low TCO, backed by built-in enterprise-level security and support for virtually
any type of data.

Data availability
High availability of mission-critical systems: SQL Server AlwaysOn
SQL Server 2014 continues to deliver on its promise of manageability through an incredible user
experience with AlwaysOn, the enhanced high-availability solution. This integrated high-availability and
disaster recovery solution provides redundancy within a data center and across data centers to help
enable fast failover of applications during planned and unplanned downtime. AlwaysOn delivers a suite
of new capabilities rolled into a single solution.

Figure 1: Unified high-availability solution

SQL Server AlwaysOn Availability Groups is a high-availability and disaster-recovery solution that
provides an enterprise-level alternative to database mirroring. Availability Groups are an integrated set of
options that include automatic and manual failover of a group of databases, support for as many as eight
secondary replicas (“secondaries”), faster failover for applications, and automatic page repair. Each
availability group is a container for a discrete set of user databases known as availability databases that
fail over together. An availability group can have many possible failover targets (secondary replicas).
Moreover, organizations can easily configure secondary replicas to support read-only access to
secondary databases and for backing up secondary databases. The addition of Availability Groups
removes the requirement of shared-disk storage such as storage area network (SAN) or network-
attached storage (NAS) for deployment of a Failover Cluster Instance.

Page 6
SQL Server AlwaysOn Failover Cluster Instances enhance SQL Server Failover Clustering and
support multisite clustering across subnets, which helps enable failover of SQL Server instances across
data centers. Faster and more predictable failover of instances is another key benefit that helps ensure
faster recovery of applications. By supporting Windows Server Cluster Shared Volumes, AlwaysOn
further improves use and management of SAN storage through increased resilience of storage failover
and avoidance of the drive-letter limitation in SAN.

SQL Server AlwaysOn Multiple, Active Secondaries enables use of as many as eight secondary
instances for running report queries (many times faster than replication) and backup operations, even in
the presence of network failures—which helps in repurposing idle hardware and improving resource
utility. It also helps to dramatically improve performance for both primary and secondary workloads
because they are no longer competing for resources.

SQL Server AlwaysOn Availability Groups Listener enables faster failover in client connections for
AlwaysOn in scenarios that employ multiple subnets. Now, client applications can achieve failover across
multiple subnets (as many as 64) almost as fast as they can achieve failover within a single subnet.
Meanwhile, the ability to set the connection from within applications to read-only (instead of read and
write) empowers organizations to control the type of workloads that run on their high-availability servers
so they can more efficiently manage their resources.

SQL Server AlwaysOn to Windows Azure Virtual Machine enables organizations to add secondary
replicas in a Windows Azure Virtual Machine through the Add Azure Replica Wizard. They can then use
this replica for disaster recovery, reporting, and backup operations. This configuration can lower capital
expenses by eliminating the need to purchase additional hardware for AlwaysOn secondaries.

Online database operations

SQL Server continues to enable organizations to achieve high availability during resource-intensive
operations. For example, the ability to rebuild online indexes in a single partition provides partition-level
control for users who need continual access to the database. This approach also requires fewer
resources (CPU and memory), so it minimizes the impact of rebuilding indexes. Specifically, the ability to
manage priorities for locking tables gives organizations greater control over the impact of maintenance
operations on running transactions—from table switching to online index rebuild operations—by allowing
database administrators to specify whether or not to terminate processes that block their ability to lock
tables.

Predictable, efficient, and flexible data backups

Recovery Advisor introduces significant enhancements to user experience through the ways in which
database administrators can restore databases by using SQL Server Management Studio. SQL Server
offers a variety of backup types, so creating the right recovery sequence for any point in time can get
tricky. To help streamline this process, SQL Server Recovery Advisor helps database administrators
create a more predictable and optimal restore sequence.

Capabilities include a visual timeline that shows the backup history of the database and the available
points in time to which the user can restore the database; algorithms that help streamline the process of
identifying the right sets of backup media to get the database back to a specific point in time; and a page
restore dialog box in SQL Server Management Studio to do page-level restores of the database.

Page 7
Backup to Azure
SQL Server enables backup and restore directly to the Windows Azure Blob service. This feature can be
used to back up SQL Server databases in an on-premises instance or in an instance of SQL Server
running a hosted environment such as Windows Azure Virtual Machine. Backup to the cloud offers
benefits such as availability, limitless geographically replicated offsite storage, and ease of transferring
data to and from the cloud. Benefits include flexible, reliable, and limitless offsite storage while providing
a great mechanism for archiving backups, and virtually no overhead to manage hardware—in addition to
cost effectiveness.

Smart Backup
Because it is built on the foundation of backup to Windows Azure, SQL Server Smart Backup provides
automatic backup policy to Windows Azure Storage that is context-aware (sensitive to workload and
throttling), uses minimum configurations (for settings such as retention period), and is able to manage
backups for the entire database instance or particular databases.

Deploy Database to a Windows Azure Virtual Machine

SQL Server provides a wizard for deploying a database to another SQL Server instance running in
Windows Azure Virtual Machine. The complexity is fully automated and does not require deep knowledge
of Windows Azure. Also, there is no performance overhead for this operation because the major factor in
defining deployment time is the size of the database itself.

SQL Server and Windows Azure Storage Integration

SQL Server provides native support for database files stored as Windows Azure blobs. This support is a
first step in moving on-premises SQL Server databases to a Windows Azure environment—to deliver an
incremental path to a database in the cloud. This feature opens up scenarios such as rapid disaster
recovery without data restore, and data encryption in the cloud with encryption keys stored on-premises.
Organizations also can move one database at a time without application changes.

Better together with Windows Server

Support for Windows Server Core

SQL Server is supported on Windows Server Core—the Windows Server edition with the smallest
footprint. Because Windows Server Core requires less maintenance and fewer OS patches, planned
downtime is greatly reduced when you run SQL Server on Windows Server Core. The percentage
reduction in patching and OS reboots can be as much as 50 to 60 percent in certain environments,
depending on the server roles that are enabled and the type of patches that are applied.2

Support for Windows Server ReFS

SQL Server supports the usage of Windows Server ReFS (Resilient File System) to provide maximum
availability, scalability, and integrity. ReFS gives organizations a cost-effective platform that maximizes
data availability, scales efficiently to very large data sets across diverse workloads, and guarantees data
integrity by providing resilience to corruption (regardless of software or hardware failures).

2
Source: “Why Is Server Core Useful,” Microsoft TechNet, http://technet.microsoft.com/en-us/library/dd184076.aspx, accessed
May 15, 2013.

Page 8
Faster live migration
Windows Server allows simultaneous migration of as many SQL Server virtual machines as you need,
which helps organizations to maintain availability of SQL Server while decreasing planned downtime.
Faster live migration also helps organizations to decrease planned downtime by allowing migration of
many SQL Server virtual machines (using priority settings) in a clustered environment, and by using as
much as 10 GB of network bandwidth.

Live migration for non-clustered virtual machines

Windows Server allows live migration of SQL Server virtual machines in a non-clustered environment
both in centrally shared and non-shared virtual machine storage scenarios. This practice helps
organizations to reduce the cost and complexity of SQL Server deployments in virtualized environments
while maintaining availability during planned downtime.

Cluster-Aware Updating
With Cluster-Aware Updating, organizations can apply updates automatically to the host operating system
or other system components in a clustered SQL Server environment while maintaining availability. This
approach can significantly help to increase SQL Server availability during the update process in both
virtualized and non-virtualized environments.

Dynamic Quorum
Windows Server Failover Clustering Dynamic Quorum enables the SQL Server AlwaysOn cluster to
dynamically adjust the number of quorum votes that are required to keep the system running. This
adjustment can simplify setup by as much as 80 percent. It also helps increase availability of a
SQL Server cluster in failover scenarios in both virtualized and non-virtualized environments—with the
ability to recalculate a quorum as needed and still maintain a working cluster.

Performance and scale

In-memory online transaction processing (OLTP)
In-memory technology for SQL Server dramatically improves the throughput and latency of SQL Server
online transaction processing (OLTP) capabilities. It is designed to meet the requirements of the most
demanding transaction processing applications and Microsoft has worked closely with a number of
companies to prove these gains. In-memory OLTP is designed on the following architectural principles:
 Optimize for main memory data access: Storage-optimized engines (such as the current OLTP
engine in SQL Server) will retain hot data in a main memory buffer pool based on frequency of
access. The data access and modification capabilities, however, are built on the viewpoint that data
may be paged in or paged out to disk at any point. With In-memory OLTP, you place tables used in
the extreme transaction processing portion of an application into memory-optimized main-memory
structures. The remaining application tables, such as reference data details or historical data, are
left in traditional storage-optimized structures. This approach lets you optimize hotspots for memory
use without having to manage multiple data engines. Main memory structures for In-Memory OLTP
eliminate the overhead and indirection of the storage-optimized view while still providing the full
atomicity, consistency, isolation, and durability (ACID) properties you expect of a database system.
 Analyze, Migrate, Report (AMR) Tool: To identify the appropriate tables to take advantage of In-
Memory OLTP, the new AMR (Analyze, Migrate, Report) Tool is integrated into the SQL Server
2014 Management Studio to assist in transition to In-Memory OLTP. It consists of a set of data
collectors, leveraging the existing Data Collection framework in SQL Server, as well as a set of
Management

Page 9
 Data Warehouse analysis reports. The reports will provide a set of recommended tables or stored
procedures for In-memory OLTP to speed the overall performance.
 Accelerate business-logic processing: With In-memory OLTP, queries and procedural logic
in procedures that are stored in T-SQL are compiled directly into machine code through
aggressive optimizations that are applied at compilation time. Consequently, the stored
procedure can be executed at the speed of native code.
 Provide frictionless scale-up: In-memory OLTP implements a highly scalable concurrency
control mechanism and uses a series of lock-free data structures to eliminate traditional locks and
latches while guaranteeing the correct transactional semantics that ensure data consistency.
 Built into SQL Server: The most impressive thing about In-memory OLTP is that it achieves
breakthrough improvement in transactional processing capabilities without requiring a separate data-
management product or a new programming model. It is still SQL Server!

In-memory data warehousing (DW): ColumnStore Index

ColumnStore Index introduces in-memory column store technology to the database engine, which makes
SQL Server the first of the major general-purpose database systems to have a true column store.
ColumnStore Index combines the VertiPaq technology that was developed in Analysis Services (and is
the foundation for PowerPivot) and a new query execution paradigm called batch processing to provide
significantly greater speed for common data warehouse queries. In test scenarios of star-join and similar
queries, organizations have experienced improvements of as much as 100 times faster than in previous
versions of SQL Server.

Figure 2: ColumnStore index in-memory columnar storage

ColumnStore Index is built on top of an existing row-level table and provides a view of the data that puts
the index on specific columns. It translates the data based on only those required columns and then
stores this view, which results in dramatic performance gains. The level of performance gains is
dependent on the data and the nature of the query.

Page 10
Database administrators also can update the Clustered ColumnStore Index online to accommodate real-
time data warehouse queries without the need to drop and recreate the index. To save disk space, they
can apply a new option called COLUMNSTORE_ARCHIVE for higher compression and storage space
savings of as much as 90 percent. Improvements to global batch aggregation also result in better
performance and more efficient processing of batch query plans when you use batch mode instead of
row mode (which consumes less memory).

Buffer pool extension

SQL Server enables improvement to query performance by allowing the use of non-volatile devices such
as solid-state drives (SSDs) to reduce SQL Server memory pressure with no risk of data loss. The
configuration is simple and can greatly improve query performance.

Query-processing enhancements

New cardinality estimator

The cardinality estimator improves the querying process and offers the following benefits:
 Consistent, predictable query performance: Different operator trees that represent the
same relation have the same cardinality estimates.
 New model for better performance: Significant changes to the model result in more
accurate cardinality estimates and better plan choices.
 Easier to support: The querying process separates querying into two steps, decision-making and
execution. It also produces tracing output to facilitate troubleshooting.

Earlier statistics invalidation

Improvements to trigger the automatic update of statistics enable better query results, because statistics
invalidation occurs earlier in SQL Server 2014. Faster and more frequent refreshes of statistics are
possible because the invalidation threshold has been set to 20 percent of a single partition.

Parallel SELECT INTO

Data-loading is significantly faster because data insertion into a table can occur in parallel through the
SELECT INTO operation.

Private cloud

Resource Governor enhancements

Resource Governor enables organizations to further ensure consistent performance for concurrent and
mixed workloads across different SQL Server applications and within private clouds. Database
administrators can define which workloads can take what percentage of performance on any given CPU,
memory, and I/O resource. Resource Governor in SQL Server brings performance scale with the
maximum number of resource pools to 64; use of minimum and maximum capacity settings in the CPU,
memory, and input/output operations per second (IOPS); and affinity of resource pools with CPU
schedulers and Non-Uniform Memory Access (NUMA) nodes. Governance of I/O resources allows
administrators to control physical I/O for users by adding a setting for maximum and minimum IOPS per
volume to Resource Governor resource pools.

Page 11
Sysprep for SQL Server
SQL Server supports the preparation of virtual machine templates through SQL Server Sysprep.
Administrators can prepare images with the desired features and then deploy them later in private
and public cloud environments. SQL Server Sysprep supports SQL Server Database Engine, SQL
Server Reporting Services, SQL Server Analysis Services, SQL Server Integration Services, and
shared features. With the addition of cluster support, SQL Server Sysprep can be used in a wider
variety of image-preparation scenarios.

Tier-1 partitioning: Scale to 15,000 partitions

SQL Server supports as many as 15,000 table partitions. This support enables large sliding window
scenarios, which means that applications such as SAP, that take tens of thousands of snapshots of data
in daily or hourly partitions, can significantly extend the length of time data is held before it’s “pushed out”
to allow for new data to enter—generally making it easier to manage these large amounts of data. This
capability also helps administrators streamline maintenance of large sets of data within file groups that
need data switched in and out according to the needs of the data warehouse.

Scalable real-world application testing: Distributed Replay

Organizations need a way to apply real-world application loads to their applications within test
environments. Previously, they could use SQL Server Profiler, which only allowed simulation of a
workload from a single computer. This limitation made it difficult to test large-scale workload simulation.

Distributed Replay helps organizations to simplify application testing and minimize errors with application
changes, configuration changes, and upgrades. This multithreaded replay utility enables simulation to test
production workload scenarios after upgrades or configuration changes—ultimately leading to protected
performance during changes. Additionally, integration with SQL Server Upgrade Assistant can help
organizations assess the impact of future SQL Server upgrades.

Reduced database size and increased performance: Data and backup compression
Many organizations want to increase speed and reliability by putting more data onto specialized disk
arrays or a SAN, but often they are prohibited by the cost of these high-end disk resources. Backup and
Data compression in SQL Server can free up space by dramatically reducing the size of databases.
Reduced data size also can increase performance. With additional space, more data can be stored on the
SAN. And because storing data on the SAN is more reliable, it also increases availability.

Additionally, SQL Server enables data compression for people who use Unicode UCS-2. This capability
enables organizations that have global language sets in their data storage to take advantage of data
compression and experience the benefits of compression.

Proactive troubleshooting and diagnostics: Performance Data Collector and Management Studio
Organizations want to proactively manage the health of their systems and quality of queries across their
environments to ensure the best possible performance. SQL Server delivers a suite of diagnostics and
tuning tools built-in and at no extra cost. Performance Data Collector allows administrators to view
SQL Server diagnostics from performance counters, dynamic management views, SQL Trace, and other
sources for baseline and historical comparisons. They can view performance data with built-in reports on
topics such as server activity, disk usage, and query activity. Additionally, SQL Server Profiler can capture
server events for real-time diagnosis, and it can correlate traces with performance counters to analyze
events and diagnose issues. Dynamic Management Views and functions that relay server state

Page 12
information help IT administrators to monitor the health of server instances, diagnose problems, and tune
performance. The Database Engine Tuning Advisor helps administrators select and create an optimal set
of indexes, indexed views, and partitions without requiring an expert understanding of the structure of the
database or the internal workings of SQL Server. Simply select the databases to tune, and Database
Engine Tuning Advisor generates indexing and partitioning recommendations.

Better together with Windows Server

High-capacity virtual machines

High-capacity virtual processors and memory in Windows Server enable organizations to deploy mission-
critical workloads by using SQL Server in a virtualized environment. A SQL Server virtual machine can
use as many as 64 virtual processors and 1 terabyte of memory. In addition, support for as many as 640
logical processors and four terabytes of memory enables deployment of mission-critical SQL Server
workloads in a non-virtualized environment.

High-density virtual machine cluster

When it is used with Windows Server, SQL Server can achieve greater cluster density for deployment in
virtualized environments, and allows as many as 8,000 SQL Server virtual machines per cluster.

High-scale cluster
Windows Server 2012 enables cluster scalability by supporting SQL Server clusters that have as many
as 64 nodes—which is four times the number supported by the previous version of Windows Server. This
increased capacity provides a range of benefits that include enhanced scalability, improved configuration
and management, and ease of maintenance for large SQL Server clusters in virtualized and non-
virtualized environments.

Network quality of service

Quality of service (QoS) in Windows Server enables administrators to enforce network bandwidth into a
network adapter for SQL Server services such as virtual machines, storage, Live Migration, and Cluster
Shared Volumes. This capability can help organizations to lower capital and operating expenses by
converging network traffic onto a single network adapter.

Network Interface Card teaming

By using Windows Server Network Interface Card (NIC) teaming to configure multiple NICs for load-
balancing, administrators can increase bandwidth for SQL Server network traffic. In addition, configuring
multiple NICs for hardware failover helps organizations to maintain SQL Server availability in virtualized
and non-virtualized environments.

Support for Server Message Block

Running SQL Server on Windows Server enables SQL Server to take advantage of Server Message
Block (SMB) file server features. SQL Server can store data files in remote shared folders that use SMB
Direct and SMB Multichannel on industry-standard network adapters. Ultimately, this capability can result
in significant storage benefits that include reduced cost, higher availability, and increased performance
for SQL Server deployments in virtualized and non-virtualized environments.

Support for Fibre Channel in Virtual Machines

SQL Server virtual machines can connect directly to Fibre Channel to support N_Port ID Virtualization
(NPIV), virtual SAN, and multipath IO (MPIO) to ensure continuous connectivity. This improvement helps

Page 13
to increase storage capacity, storage compatibility, and overall performance for SQL Server deployments
in virtualized environments.

Storage Pools
Organizations have an opportunity to lower costs by using industry-standard storage in SQL Server
deployments for non-virtualized environments—and in some cases, even replace expensive SAN
solutions. Storage Pools can improve the flexibility of SQL Server storage with resilient storage (mirroring
and parity) and multitenancy isolation (ACLs).

Organizational security and compliance

Secure by default: Lowering vulnerability
Microsoft and the SQL Server team take security seriously. More than 10 years ago, Microsoft
implemented the Trustworthy Computing initiative. The initiative requires SQL Server engineers to take
regular security training and carry that responsibility for security across their job duties, regardless of the
group in which they reside. This company-wide discipline to protect security and privacy was developed
to create software that is secure by design—and to reduce by default the overall risks related to security.
To that end, according to the National Institute of Standards and Technology (NIST) public security
board, SQL Server reportedly has the lowest number of security vulnerabilities across the major database
vendors. In addition, SQL Server has been deemed “the most secure database” by the Information
Technology Industry Council (ITIC).3

Built-in tools for enabling compliance: SQL Server audit enhancements

Database auditing is built into SQL Server to make auditing easy because it is continually available, and
to help organizations audit database activities, including database reads, with minimal impact to
performance. As compliance policies get increasingly tighter, organizations can use built-in tools such as
the following:
 SQL Audit (all editions): Enables organizations to extend the benefits of SQL Server Audit
from Enterprise edition to all SQL Server editions. This extensibility allows for more thorough
auditing practices across SQL Server databases, and it enables standardization, better
performance, and richer features.
 User-Defined Audit: Allows the middle-tier application to write custom events into the audit
log, which enables more flexibility to store audit information.
 Audit Filtering: Provides greater flexibility to filter unwanted events in an audit log.
 Audit Resilience: Gives the ability to recover auditing data from temporary file and network issues
to help ensure that audit logs are not lost during failover.

Restricted access to data at the row level: Label security

Many industries and government agencies require the ability to restrict access to data at the row and cell
levels, to ensure that only authorized people have access to the data. With SQL Server, organizations
can enact this detailed approach to security by using the Microsoft toolkit SQL Server Label Security at
no additional cost. Organizations can use this toolkit to support critical scenarios or niches within their
database environments—without having to purchase costly add-ons. The toolkit, which is developed and
maintained by Microsoft engineers, is available on http:///sqlserverlst.codeplex.com.

3
Source: Information Technology Intelligence Corp. (ITIC), SQL Server Delivers Industry-Leading Security, September 2012.

Page 14
Controlled access to data in motion: Advanced security
SQL Server allows organizations to encrypt data when it is stored on a disk and decrypt it when it is read
into memory. Therefore, organizations do not have to make changes to their applications for SQL Server
to secure their data. Because encryption is built into the database engine, it is transparent to applications
and users—and it is included in SQL Server Enterprise edition. Encryption also protects database
backups by automatically encrypting the backups.

Additionally, extensible key management works with transparent data encryption to store encryption keys
outside of the database. With extensible key management, organizations can use a hardware device or
a third-party encryption tool to create encryption keys. Storing the keys separately from the encrypted
data makes it even harder for unauthorized users to gain access to encrypted data.

Controlled access to data for administrators: User-Defined Server Roles

User-Defined Server Roles increases flexibility and manageability, and it facilitates compliance through
clearer separation of duties. It allows creation of server roles to suit different organizations that separate
administrative duties according to roles. Roles also can be nested to allow more flexibility in mapping to
hierarchical structures in organizations. User-Defined Server Roles also helps prevent organizations from
using sysadmin for database administration. For example, a special database administration role can be
created for common database administration without the ability to access user data.

Backup Encryption
Encryption is a straightforward way to increase security of the data. Encryption of backups is the way to
increase security of backups stored separate from the instance, and probably in the different
environment. Separate encryption settings for backup allow configuring backup encryption differently
from database encryption and benefit from the both. In case database is not encrypted backup
compression might be effectively used together with encryption improving as security, same as storage
and transfer costs.

Controlled access to data across business intelligence tools: Microsoft SharePoint and Microsoft
Active Directory
As Microsoft continues to deliver business intelligence tools that are used by a broader set of users,
security concerns also increase because of broader implications if security is compromised. SQL Server
helps organizations secure end-user data analytics with built-in IT controls, including new SharePoint and
Active Directory security models for end-user reports that are published and shared in SharePoint.
Enhanced security models provide control at row and column levels.

Virtually any data, built-in

Embracing unstructured data through support for complex data types
With the dramatic shift from structured to complex data types, the new mission critical requires support
for complex data types to be built-in, not added on at extra cost and effort. SQL Server supports a
growing number of types and volumes of complex data with FILESTREAM, Remote Blob Storage, and
Spatial support—enhancements on top of the already robust and built-in foundation that extend beyond
relational capabilities. This support allows organizations to build rich and innovative applications without
paying extra premiums.

SQL Server FileTable builds on FILESTREAM to bring Win32 namespace support and application
compatibility to the file data stored in SQL Server. Countless applications maintain their data in two

Page 15
“worlds:” unstructured (documents, media files, and other unstructured data in file servers) and structured
(related, structured metadata in relational systems). FileTable lowers the barrier to entry for organizations
who have files on servers that currently run Win32 applications, while reducing the effort caused by
maintaining two disparate systems and keeping them in sync.

High availability for unstructured data

With SQL Server, complex data types are handled with the same attention as common data types.
Organizations can use FILESTREAM to store and manage complex data in a variety of ways, as if it were
part of the database. Additionally, with SQL Server, organizations can enjoy the high-availability benefits
of AlwaysOn for complex data managed through FILESTREAM—even when they take advantage of
Remote Blog Storage and FileTable.

Seamless connection and analysis of Big Data through Hadoop connectors

Hadoop connectors for SQL Server and SQL Server Parallel Data Warehouse are available for download
to organizations that have licenses for SQL Server and Parallel Data Warehouse. These connectors
enable bidirectional data movement across SQL Server and Hadoop, so users can work effectively with
both structured and unstructured data. Additionally, people can use the leading business intelligence
platform from Microsoft to perform analysis on Hadoop data sets. Users can gain access to and create
mashups of Hadoop data sets by using familiar productivity tools such as Microsoft Excel and award-
winning business intelligence clients such as PowerPivot and Power View to perform analysis in an
immersive and interactive way.

Seamless connection to a variety of platforms through greater interoperability Realistically,

many organizations own and manage heterogeneous infrastructures. They need interoperable tools
and systems to maximize their existing investments and evolve on their own terms. SQL Server helps
organizations extend heterogeneous environments by connecting to SQL Server and Windows Azure
SQL Database applications, and provides support for additional industry-standard APIs across a
variety of platforms. By using SQL Server, organizations can maximize existing investments while they
modernize legacy mission-critical workloads on the SQL Server data platform.

Figure 3: Interoperability across platforms and standards

Interoperability with SQL Server includes the following:

 Microsoft Driver for PHP for SQL Server: Designed to enable reliable, scalable integration
with SQL Server for PHP applications that are deployed on the Windows platform.
 Connectivity for Java: Provides secure and highly available connectivity from Java applications
to SQL Server for enterprise-level organizations.
 Microsoft JDBC Driver for Linux and UNIX: Provides connectivity to Linux and UNIX, which
allows organizations that run applications and workloads on legacy platforms to more easily migrate
to
SQL Server with minimal effort.

Page 16
Comparing the cost of mission-critical capabilities
In a technology landscape where organizations expect vendors to intuitively know what “mission critical”
means and to provide easy and cost-effective solutions, Microsoft answers these expectations with
enterprise-class tools and abilities that are built into database technology—without the need to purchase
costly add-ons. SQL Server delivers the mission-critical capabilities required by organizations to compete
in a dynamic digital world. The features discussed in this paper are all included in SQL Server Enterprise
Edition and they don’t require costly options to deliver a complete and modern database solution.

Expecting the fundamentals

Many legacy vendors deliver limited functionality in their premium editions and only provide mission-
critical requirements through additional options or feature packs. Examples of additional options are
security features, high availability, performance, and spatial capabilities. Organizations have evolved—
and it’s no longer optional for vendors to ensure enterprise-grade security features, availability,
performance or support for complex data types, to name a few.

Just like home buyers expect a roof, windows, and doors to be included in their purchase, an
organization can expect an enterprise-class database to include built-in availability, performance, and
security features. Figure 4 shows the difference between two similar database solutions, Microsoft SQL
Server and Oracle Database, with the budget impact of adding options to arrive at a similar end state.

Figure 4: SQL Server and Oracle Database compared

Understanding options
Table 1 highlights the options required across the major database management system (DBMS) vendors
to meet mission-critical needs in modern organizations. What used to be optional is more often required
by organizations to meet the new standard in mission-critical operations. It is easy to see how achieving
mission-critical readiness by adding options or feature packs can dramatically change the total cost of a
database solution.

Page 17
Page 18
Table 1: Comparison of mission-critical solutions from Microsoft and Oracle

Microsoft SQL Server Oracle Database

(all options not shown)
Enterprise Edition $27,496 $95,000
base license (includes
1-year support)
Data availability Included $10,000 (Active Data Guard)
$20,000 Total
Performance and scale Included $11,500 (Advanced Compression)
$23,000 (In-Memory Cache)
$11,500 (Partitioning)
$11,500 (Real Application Testing)
$5,000 (Diagnostics Pack)
$5,000 (Tuning Pack)
$135,000 Total
Enterprise security Included $11,500 (Advanced Security)
Free Download $11,500 (Label Security)
$46,000 Total
Any data, built-in Included $17,500 (Spatial & Graph)
$35,000 Total
Total cost $27,496 $331,000

Note Microsoft prices are based on estimated retail price. All Microsoft and Oracle prices are per-processor (based on a quad core
Intel Xeon processor) database pricing for purchases within the United States and are in United States dollars. Pricing is based on
information available on vendor websites. Oracle prices are based on the Oracle Technology Global Price List, March 15, 2013. IBM
prices may vary.

In addition to its mission-critical database functionality, SQL Server Enterprise includes a range of
capabilities for data integration, data management, data cleansing, and end-to-end business intelligence.
According to the Gartner Magic Quadrant for Business Intelligence and Analytics Platforms,4 Microsoft is
positioned as a leader in helping organizations to enable broad end-user insight and productivity—
balanced with IT oversight through managed self-service business intelligence tools that work both
standalone and within Microsoft SharePoint. SQL Server delivers access to these industry-leading
business intelligence capabilities without requiring costly add-ons. With SQL Server, the business
intelligence tools are built into the base Enterprise license and are also available in the new Business
Intelligence edition. Organizations also can increase cost-savings through built-in data integration,
management, and cleansing tools. These tools enable data quality managers to easily cleanse and
manage data through SQL Server Integration Services, Master Data Management, and Data Quality
Services. Similar business intelligence and data management tools with other vendors can add up to
hundreds of thousands of dollars in additional costs.

4
Source: Gartner, Magic Quadrant for Business Intelligence and Analytics Platforms, February 5, 2013.

Page 19
Conclusion
SQL Server delivers a new standard in enabling mission-critical operations—with true enterprise-class
availability, performance, and security features built into the solution. Integrated high-availability solutions
enable faster failover and more reliable backups—and they are easier to configure, maintain, and
monitor, which helps organizations reduce TCO. SQL Server also delivers mission-critical performance
and scale, with predictable performance across server activities including complex queries, data
integration, and analysis. Because SQL Server is designed to security standards, it has minimal total
surface area and database software that is inherently more secure. Enhanced security, combined with
built-in, easy-to-use tools and controlled data access, helps organizations meet strict compliance
policies. SQL Server supports complex data types and non-traditional data sources, and it handles them
with the same attention—so organizations experience seamless support for a variety of platforms and
heterogeneous environments. Finally, SQL Server delivers mission-critical capabilities at low TCO—with
full enterprise capabilities that are built into the solution, not provided as costly add-ons. Ultimately,
organizations can rely on a comprehensive, integrated solution that helps to contain costs and manage
compliance requirements, while meeting the demands of the evolving digital world.

For more information, see the SQL Server website at http://www.microsoft.com/sqlserver/.

Page 20