1. SQL Server 2017 Editions
Find the right solution for your business

2. Reason over any data, anywhere
The modern data estate
Cloud
HYBRID
On-premises
Private cloud
Operational databases
Operational databases
Data warehouses
Data warehouses
The modern data estate spans across on-premises and the cloud, where operational databases, data warehouse, and data lake combine to drive insights from data. In order to innovate, the modern data estate should reason over any data, flexibility of choice, and security and privacy.
Data lakes
Data lakes
Reason over any data, anywhere
Flexibility of choice
Security and privacy

3. 1/10th the cost of Oracle
SQL Server 2017
Industry-leading performance and security NOW ON LINUX and DOCKER
Choice of platform and language
Industry-leading performance 
Most secure over the last 7 years
Only commercial DB with AI built-in
End-to-end mobile BI on any device
Microsoft
Tableau
Oracle
$120
$480
$2,230 R
1/10
Self-service BI per user
T-SQL
Java
C/C++
C#/VB.NET
PHP
Node.js
Python
Ruby
#1 TPC-H performance 1TB, 10TB, 30TB
#1 TPC-E performance
#1 price/performance
#1 price/performance in TPC-H non-clustered as of 9/1/
#1 TPC-H non-clustered benchmark as of 9/1/
#1 TPC-E performance as of 9/1/
R and Python + in-memory at massive scale
Native T-SQL scoring
A fraction of the cost
In-memory across all workloads
Most consistent data platform
Private cloud
Public cloud

4. SQL SERVER 2017 ACROSS EDITIONS
Making innovation more accessible to all applications
Delivers common programming surface across editions—no application re-write
Advanced featured across editions
SQL Server 2017 provides advanced features, regardless of which version you select. Enjoy common programming surface across all editions with no app re-write.
Up to 30x faster transactions, 100x faster queries with in-memory performance
Real-time operational analytics without impacting performance
Only data solution to encrypt your data at rest and in motion
Connect your relational data to big data with PolyBase
Unparalleled choice for developer tools and languages

5. SQL SERVER 2017 EDITIONS 8/22/2018 Enterprise Standard Express
Common programming surface area - develop once and scale across editions
MISSION CRITICAL IN-MEMORY PERFORMANCE AND SCALE, SECURITY AND HIGH AVAILABILITY
Mission critical high availability on Windows and Linux
Enhanced in-memory performance
Faster performance with Adaptive Query Processing
Unparalleled data security
PB scale data warehousing
End-to-end mobile BI with rich visualizations on all major platforms
In-database advanced analytics built-in at scale with R and Python
Enhanced hybrid scenarios including Stretch Database, HA, DR and backup
Software Assurance benefits include unlimited virtualization, Machine Learning Server for Hadoop, and Power BI Report Server
Standard
FULLY FEATURED DATABASE FOR MID- TIER APPLICATIONS AND DATA MARTS
End-to-end database security with Always Encrypted
Enhanced in-memory performance for all workloads
Basic reporting
Basic analytics
Hybrid scenarios: Stretch Database, backup
Express
SMALL-SCALE APPLICATIONS
Development and management tools
Easy backup and restore to Microsoft Azure
Free to use
This is a simplified 1-slide version for customers to show the all-up positioning of these four main SKUs.
Enterprise: Comprehensive, high-end capabilities for demanding database and business intelligence requirements.
Standard: Core data management and business intelligence capabilities for non-critical workloads with minimal IT resources.
Express: Easy to get started, free to use for redistribution and embedding.
Developer: Free version where customers can build, test, and demo apps in non-production environments
Now we’ll look at features in detail with a focus on Standard and Enterprise only.
Developer
DEVELOPMENT AND TESTING  Build, test, and demo apps in non-production environments  Free to use  All Enterprise Edition features available
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. 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 provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

6. Business Intelligence
SQL Server 2017 features by edition
Standard
Enterprise
OLTP Performance
Maximum number of cores
24 cores
Unlimited
Maximum memory utilized per instance
128 GB
OS Max
Maximum size
524 PB
Advanced OLTP (In-memory OLTP*, Operational analytics) 
Basic high availability (2-node single database failover, non-readable secondary)
Manageability (Management Studio, Policy-Based Management)
Enterprise data management (Master Data Services, Data Quality Services)
Advanced HA (Always On - multi-node, multi-db failover, readable secondaries)
Basic Adaptive Query Processing (Interleaved execution)
NEW*
Advanced Adaptive Query Processing (Batch mode memory grant feedback, Batch mode adaptive joins), Automatic Plan Correction
Security
Basic security (Always Encrypted, Row-level security, data masking, basic auditing, separation of duties)
Advanced security (Transparent Data Encryption)
Data Warehousing
Advanced data integration (Fuzzy grouping and look ups, change data capture)
Data warehousing (In-Memory ColumnStore, Partitioning)
T-SQL query across relational and Hadoop data with PolyBase**
Business Intelligence
Programmability & Developer Tools (T-SQL, CLR, Data Types, FileTable, JSON, graph data support)
Basic data integration (SSIS, built-in connectors)
Basic reporting & analytics*
Basic Corporate Business Intelligence (Multi-dimensional models, Basic tabular model, enhanced connectors, new transformations, object-level security, ragged hierarchies)*
Mobile BI*
Advanced Corporate Business Intelligence (Advanced tabular model, Direct query, in-memory analytics, advanced data mining)*
Advanced Analytics
Basic R and Python integration (Connectivity to R Open and Python, Limited parallelism)*
Advanced R and Python integration (Ability to run on GPUs and full parallelism through Machine Learning Services)*
Hybrid Cloud
Stretch Database*
General approach is to scale features down from EE to SE with throttles such that features are programmable in SE but EE is required for scale and performance:
Tabular Models in SE with 16GB memory cap per instance
In-memory OLTP in SE with 64MB memory restriction at the instance level
In-DB-Analytics allow developers to execute RRE-functions outside EE with close-to-RRO limitations (perf & dataset size)
Security, encryption story is fractured: TDE moves from EE to SE; Always Encrypted is EE
1-page overview of features by edition with new features highlighted.
Features in blue are newly available.
*In-memory OLTP subject to memory limits in Standard Edition **Available for Windows Server only

7. DRAMATICALLY SIMPLIFY HA&DR
Enterprise
MISSION CRITICAL HA + HYBRID DR
Multiple node failover clustering on Windows and Linux NEW* (3 synchronous replicas and up to 9 total replicas)
Multiple database failover as a group
Fully readable secondaries for backup or reporting operations
Distributed Availability Groups NEW* provide DR resiliency and enhance read scale by increasing the number of readable secondaries across geographies
Read scale-out without cluster dependency NEW*
No domain join required with WS 2016
DRAMATICALLY SIMPLIFY HA&DR
High Availability with the 9s you need for mission critical workloads
Enterprise Edition availability with enhanced Always On
Cost effective to run backups and scale BI reporting
Hybrid DR with Azure
Azure VMs
Asynchronous replicas
High Availability
Standard
BASIC HA
Two node failover (1 active, 1 passive)
Single database failover to non-readable secondary
No domain join required with WS 2016
Synchronous replicas on Windows
Primary
Mission critical availability on any platform
In preparation for the release of SQL Server v.Next, we are enabling the same High Availability (HA) and Disaster Recovery (DR) solutions on all platforms supported by SQL Server, including Windows and Linux. Always On Availability Groups is SQL Server’s flagship solution for HA and DR. Microsoft has released a preview of Always On Availability Groups for Linux in SQL Server v.Next Community Technology Preview (CTP) 1.3.
SQL Server Always On availability groups can have up to eight readable secondary replicas. Each of these secondary replicas can have their own replicas as well. When daisy chained together, these readable replicas can create massive scale-out for analytics workloads. This scale-out scenario enables you to replicate around the globe, keeping read replicas close to your Business Analytics users. It’s of particularly big interest to users with large data warehouse implementations. And, it’s also easy to set up.
In fact, you can now create availability groups that span Windows and Linux nodes, and scale out your analytics workloads across multiple operating systems.
New flexibility to do HA without Windows Server fail over clustering
Fail-over clustering with Pacemaker and more through integration scripts and guides
Always On availability groups with automatic fail-over, listener, synchronous replication, read-only secondaries
Shared disk failover clusters
Backup and restore: .bak, .bacpac, and .dacpac
Log shipping
Synchronous replicas on Linux
Standard Edition High Availability
High Availability
Non-readable synchronous replica
Primary

8. 1M prediction per second
Enterprise
R and Python NEW* built-in to your T-SQL
Enhanced R and Python NEW* APIs with full parallelism and no memory limits for scale/performance
The ability to run on GPUs NEW*
Built-in In-memory Advanced Analytics
Advanced tabular model
Direct query
Advanced data mining
SSDT in Visual Studio
Native Scoring with T-SQL designed for high performance concurrent scoring scenarios
ADVANCED ANALYTICS
Bring Advanced Analytics to your data
R and Python built-in to SQL Server
Age
Original balance
Interest rate
Loan remaining months
Credit score
Business user
Standard
Connectivity to Microsoft R Open and open-source Python analytics NEW*
R and PythonNEW* APIs with serial execution and memory limitations (2 cores & memory of the host machine)
Native Scoring with T-SQL designed for high performance concurrent scoring scenarios R
Store predictions
Visualize
The first commercial database with AI built-in
In the past, a common application pattern was to create statistical and analytical models outside the database, in the application layer or in specialty statistics tools. Our approach is to bring intelligence to where the data lives. We are bringing the ability to run advanced analytics models in a performant multi-threaded, distributed way at scale directly in the database. This includes the Microsoft Machine Learning algorithms grown from Microsoft’s leading AI research made available to you both on-premises and in the cloud – unlike anyone else.
Another new, key feature enhancement in CTP 2.0 of SQL Server 2017 is the ability to run the Python language in-database to scale and accelerate machine learning, predictive analytics and data science scripts. The new capability, called Microsoft Machine Learning Services, enables Python scripts to be run directly within the database server, or to be embedded into T-SQL scripts, where they can be easily deployed to the database as stored procedures and easily called from SQL client applications by stored procedure call. SQL Server 2017 will also extend Python’s performance and scale by providing a selection of parallelized algorithms that accelerate data transforms, statistical tests and analytics algorithms. This functionality and the ability to run R in-database and at scale are only available on Windows Server operating system at this time.
By adding Python support in addition to R and adding real-time scoring capabilities, now you can run machine learning models directly in SQL Server to eliminate the need to move data, increase efficiency and help uncover new insights.
With Python support, customers also gain the ability to work with the latest deep learning frameworks like Cognitive Toolkit and Tensor Flow to run deep learning models against data that lives in the SQL Server database.
Definitions:
GPU – offload some of heavy compute to the graphics processing unit (GPU). As a result, applications simply run faster. This is particularly applicable to deep learning, analytics, and engineering applications.
Native scoring with T-SQL - Microsoft created an extensibility framework that allows R scripts to be executed from T-SQL. This framework supports any operation you might perform in R, ranging from simple functions to training complex machine learning models. However, the dual-process architecture requires invoking an external R process for every call, regardless of the complexity of the operation. If you are loading a pre-trained model from a table and scoring against it on data already in SQL Server, the overhead of calling the external R process represents an unnecessary performance cost. Scoring is a two-step process. First, you specify a pre-trained model to load from a table. Second, pass new input data to the function, to generate prediction values (or scores). The input can be either tabular or single rows. You can choose to output a single column value representing a probability, or you might output several values, such as a confidence interval, error, or other useful complement to the prediction. <
Columnstore
In-Memory OLTP
Power BI dashboard
1M prediction per second
Prepare for analytics

9. ANALYSIS SERVICES Enterprise Standard
Multi-dimensional models
Advanced tabular model
Enhanced connectors, new transformations, object-level security, ragged hierarchies NEW*
OS max memory capacity per instance to support larger models
Real-time access without moving data with Direct Query storage mode
KPIs and translations
DAX calculations, queries and MDX queries
Object-level security NEW* and row-level security
Faster performance through parallel processing of partitions and in-memory storage mode
ANALYSIS SERVICES
Enterprise grade analytics engine – on-premises and in the cloud
Modern data sources
Traditional data sources My
10 01
Facebook
MailChimp
MySQL
Azure Blob
RedShift
Files
SQL
Oracle
SAP
Others
Analysis Services
BI semantic model
Standard
Multi-dimensional models
Basic tabular model
Enhanced connectors, new transformations, ragged hierarchies NEW*
16 GB memory capacity per instance
KPIs and translations
DAX calculations, queries and MDX queries
Object-level security NEW* and row-level security
Faster performance through parallel processing of partitions and in-memory storage mode
Data modeling
In-memory cache
Security
Business logic & metrics
Lifecycle management
Analyze diverse data with powerful modelling in SQL Server Analysis Services
Modern data connectivity & transformation
Modern connectors NEW* for big data sources, additional databases, Azure stores, and diverse modern data sources
Fast time to insights with rich data transformations and mashups NEW*
In-memory performance
Data mashup transformations
Drill-through & ragged hierarchies
Improved scalability on enterprise hardware & DirectQuery performance enhancements
All of these enhancements apply to tabular models
For all models, tabular and multi-dimensional:
SSDT with local AS engine, no SQL Server dependency
Support for Visual Studio “Dev15”
Definitions:
Ragged hierarchies - A ragged hierarchy is a user-defined hierarchy that has an uneven number of levels. Common examples include an organizational chart where a high-level manager has both departmental managers and non-managers as direct reports, or geographic hierarchies composed of Country-Region-City, where some cities lack a parent State or Province, such as Washington D.C., Vatican City, or New Delhi <
Object level security – secure table and column names in addition to the data within them
Enhanced connectors – support for Tabular Object Model APIs and TMSL scripting
Visualize
Excel
SSRS
Power BI Report Server
Power BI
Third party tools

10. DATA WAREHOUSING AND DATA MARTS
Enterprise
In-Memory Columnstore
Partitioning, compression, and change data capture
Petabyte Enterprise Data Warehouse scale
PolyBase in scale-out configuration—head and compute nodes
Scale out, read-only Analysis Services configuration
Global batch aggregation
Star query join optimization
Advanced data mining
DATA WAREHOUSING AND DATA MARTS
Query across relational and non-relational data
PolyBase
T-SQL query
Quote:
$658.39
Standard
Data Marts and OLAP Cubes
PolyBase—compute node only
In-Memory Columnstore
Partitioning, compression, and change data capture
SQL Server
Hadoop
Petabyte-scale Data Warehousing
1TB benchmark - we are excited to announce a world record in the TPC-H 1TB data warehousing workload (non-clustered). The benchmark was achieved with SQL Server 2017 on Red Hat Enterprise Linux and HPE Prolliant server hardware, beating SQL Server 2016 on the same hardware handily.  This is just the first of many anticipated performance benchmarks for SQL Server 2017 on Linux and Windows, demonstrating SQL Server’s industry leading performance. 
Recent years have seen a data explosion with accompanying challenges in data storage and information retrieval. Web sites are streaming data directly into corporate databases at a rate unthinkable just a few years ago. Databases are swelling to sizes unmanageable with currently technology. Extracting information from this massive amount of data is getting more complicated as the questions get more and more sophisticated. Organizations are missing opportunities and wasting effort.
Microsoft has stepped up to the challenge by introducing the Data Warehouse Fast Track Reference Architecture, a set of guidelines that partners can use to help customers build medium-to-large data warehouse solutions based on SQL Server 2016 Enterprise edition and well-tuned configurations from certified hardware vendors. Fast Track is an SMP data warehouse option that has a current on-premises data compute capacity range that seamlessly scales from 5 TB to 145 TB, with a data storage capacity that reaches past 1 PB. Microsoft is constantly pushing the capacity envelope, and the Data Warehouse Fast Track program is rapidly extending this range with the publication of new benchmarks.
This mature deployment, available from certified vendors, works best when matched to customer data and application requirements. The Fast Track program has many administrative, operational and programming capabilities and is a recommended best experience for the smaller and mid-size enterprise data warehouse.
Customers can also run large SQL Server data warehouse in the cloud on an Azure VM with daisy-chained storage, or get scale-up MPP (Massive Parallel Processing) from our data warehouse-as-a-service, Azure SQL Data Warehouse.
NAME
DOB
STATE
Denny Usher
11/13/58 WA
Gina Burch
04/29/76

11. UNPARALLELED SECURITY FOR ALL YOUR APPS
Enterprise
Transparent data encryption (TDE)
Always Encrypted
Row-level security
Data masking
Fine-grained auditing
Separation of duties
UNPARALLELED SECURITY FOR ALL YOUR APPS
Protect data at rest and in motion
Client side
Server side
Always Encrypted
SQL
Query
Enhanced SQL Server
Library
Standard
Always Encrypted
Row-level security
Data masking
Fine-grained auditing
Separation of duties
Protect your data at all times on Windows and Linux
Most secure platform, with the least vulnerabilities in the NIST vulnerability database
We have technologies available both at the infrastructure and the database level that have landed us in this number one spot.
We give you great tools to protect your data, control access and monitor activity.
Encrypt data at rest and in motion with Always Encrypted on new driver libraries
We also added better access control. Windows authentication is improving significantly in Windows Server 2016, but there is also row level security and dynamic data masking for controlling access.
We can protect data at rest and in motion with this new technology, called Always Encrypted. Microsoft research worked on this technology for over three years, and we brought it into this product for the first time in SQL Server 2016.
What’s especially compelling is that this is client side encryption technology. It does basically all of the heavy lifting on the client side. To roll it out, all the customer needs is .Net Framework 4.6 or higher, or an updated connection driver like the SQL Server ODBC and JDBC drivers. As a bonus, it has a silent install which makes deploying it across an enterprise easy as it can be done silently.
We talk about protecting data at rest and in motion without impacting database performance and the second line is the key. While this may look like something Oracle has, we hear from Oracle customers that they don’t use it because it kills database performance. This is the beauty of Always Encrypted.
Another great benefit to point out to you is the location of the master key. It’s not with SQL Server, instead it sits in a trusted area with the customer. This works with Azure key vault as well as other solutions.
Building the slide animation shows how the technology works: the query lookup is converted to cyphertext so that the appropriate row can be retrieved without decrypting the cyphertext on the SQL Server side. The decryption of the cyphertext only happens in the trusted area on the client side.
At customer events, we often demo a “man in the middle” attack to showcase that the social security was still in ciphertext during the attack on the wire and even in the memory buffer pool.
This technology is one of the heroes of 2016, and now 2017 and the biggest selling point is that we can do this without impacting database performance because all of the heavy lifting is on the client side.
With any kind of a hybrid scenario that includes sensitive data, if you want to know how we can ensure that it’s secure if SQL Server gets hacked, Always Encrypted is the answer.
Conceal sensitive information with Dynamic Data Masking
Control access to database rows based on user characteristics with Row-Level Security
File-level protection with Transparent Data Encryption
In SQL Server, we improved auditing capabilities across all these layers so you can monitor and track threats.
Data set
CIPHERTEXT
Customer
Credit card #
Exp.
Tim Irish
7/19
Denny Usher
5/17
Alicia Hodge
4/18
Credit card #
1x7fg655se2e
0x7ff654ae6d
0y8fj754ea2c
Customer
Credit card #
Exp.
Denny Usher
0x7ff654ae6d
5/17
Column master key
Column encryption key

12. up to 128GB memory and 24 cores up to 24TB memory and 384 cores*
SQL Server Enterprise Edition
Powered for mission critical workloads
All of the features of SQL Server, plus additional advantages
Unparalleled performance at scale
Mission-critical availability
Mobile BI on any device
Unmatched data security
Experience maximum performance, cores and memory
Maximize uptime with multiple readable secondaries
Deliver self-service BI at a fraction of the cost of competitors
Encrypt data at rest with Transparent Data Encryption
This slide highlights key SQL Server Enterprise Edition differentiation
SQL Server Enterprise Edition running on 24 TB of memory and 384 cores occurred in a server in the Fort Collins HP laboratory.
Standard Edition
up to 128GB memory and 24 cores
Enterprise Edition
up to 24TB memory and 384 cores* HA
Multiple readable secondaries
Primary
*Enterprise Edition does not limit maximum memory or cores

13. SOFTWARE ASSURANCE BENEFITS ACROSS EDITIONS
SQL Server 2017 Editions
Standard
Enterprise
Next version rights
Stay up to date with the latest features 
License Mobility to shared third party servers
Modernize to the cloud with existing licenses
Fail-Over Servers for high availability
Take advantage of one passive secondary server for no additional licensing cost
Unlimited virtualization
License all physical cores on the server and enable unlimited virtual machine deployments
Running Machine Learning Server for Hadoop
Get access to quarterly updates to the advanced analytics stack
Running Power BI Report Server
Generate data visualizations on premises with Power BI Report Server
SA benefits vary across editions.

14. Appendix

15. Flexible, reliable data management SQL Server on the platform of your choice
Windows
Linux
Support for RedHat Enterprise Linux (RHEL), Ubuntu, and SUSE Enterprise Linux (SLES)
Linux and Windows Docker containers
Windows Server / Windows 10
Package-based installation: Yum Install, Apt-Get, and Zypper
Linux/Windows container
Last but not least, customers need flexibility when it comes to the choice of platform, programming languages & data infrastructure to get from the most from their data.
Why? In most IT environments, platforms, technologies and skills are as diverse as they have ever been, the data platform of the future needs to you to build intelligent applications on any data, any platform, any language on premises and in the cloud.
SQL Server manages your data, across platforms, with any skills, on-premises & cloud
Our goal is to meet you where you are with on any platform, anywhere with the tools and languages of your choice.
SQL now has support for Windows, Linux & Docker Containers.
It allows you to leverage the language of your choice for advanced analytics – R & Python.

16. What’s available in SQL Server on Linux
Windows
Linux
Editions
Developer, Express, Web, Standard, Enterprise 
Services
Database Engine, Integration Services
Master Data Services, Data Quality Services
Mission critical performance
Maximum number of cores
Unlimited
Maximum memory utilized per instance
12 TB
Maximum database size
524 PB
Basic OLTP (Basic In-Memory OLTP, Basic operational analytics)
Advanced OLTP (Advanced In-Memory OLTP, Advanced operational analytics, adaptive query processing)
Basic high availability (2-node single database failover, non-readable secondary)
Advanced HA (Always On - multi-node, multi-db failover, readable secondaries)
Security
Basic security (Basic auditing, Row-level security, Data masking, Always Encrypted)
Advanced security (Transparent Data Encryption)
Data warehousing
PolyBase
Basic data warehousing/data marts (Basic In-Memory ColumnStore, Partitioning, Compression)
Advanced data warehousing (Advanced In-Memory ColumnStore)
Advanced data integration (Fuzzy grouping and look ups)
Tools
Windows ecosystem: Full-fidelity Management & Dev Tool (SSMS & SSDT), command line tools
Linux/OSX/Windows ecosystem: Dev tools (VS Code), DB Admin GUI tool, command line tools
Developer
Programmability (T-SQL, CLR, Data Types, JSON, Graph)
Windows Filesystem Integration - FileTable
BI & Advanced Analytics
Corporate Business Intelligence (Analysis Services, Reporting Services, Multi-dimensional models, Basic tabular model)
Machine Learning Services (R and Python integration)
Hybrid cloud
Stretch Database
What’s available in SQL Server on Linux
Slide highlights the available features on Linux vs Windows

17. WHAT’S NEW IN SQL SERVER 2017
Choice of platform
OLTP performance
Business Intelligence
AI built-in
Unlimited cores, OS max memory
Enterprise
Linux support
Container support
Advanced Adaptive Query Processing (Batch mode memory grant feedback, Batch mode adaptive joins)
Automatic Plan Correction
Enhanced connectors, new transformations, object-level security, ragged hierarchies*
Graph data support
Access to Power BI Report Server, a software assurance benefit
GPUs and full parallelism through Machine Learning Services*
Machine Learning for Hadoop/Spark and Machine Learning for Linux, a software assurance benefit
24 Cores, 128 GB memory
Standard
Basic Adaptive Query Processing (Interleaved execution)
Connectivity to Python, limited parallelism*
What’s new in SQL Server 2017
Adaptive Query Processing: adaptive query processing features that you can use to improve query performance in SQL Server and Azure SQL Database, including:
Batch mode memory grant feedback, Batch mode adaptive join, and Interleaved execution. Sometimes the plan chosen by the query optimizer is not optimal for a variety of reasons. For example, the estimated number of rows flowing through the query plan may be incorrect. The estimated costs help determine which plan gets selected for use in execution. If cardinality estimates are incorrect, the original plan is still used despite the poor original assumptions.
<
Automatic Plan Correction: Automatic plan correction is a new automatic tuning feature in SQL Server 2017 that identifies SQL query plans that are worse than previous one, and fix performance issues by applying previous good plan instead of the regressed one. SQL Server can use different strategies (or SQL plans) to execute a T-SQL query. SQL Server will analyze possible plans that can be used to execute a T-SQL query and choose the optimal plan. The plans for most of the successfully executed queries are cached and reused when the same query is executed. The plan is retained in the cache until SQL Database engine decides to recompile the plan and find a new one (e.g. when statistics change, index is added or removed, etc.)
In some cases, new plan that is chosen might not be better than the previous plans.
<
Ragged hierarchies - A ragged hierarchy is a user-defined hierarchy that has an uneven number of levels. Common examples include an organizational chart where a high-level manager has both departmental managers and non-managers as direct reports, or geographic hierarchies composed of Country-Region-City, where some cities lack a parent State or Province, such as Washington D.C., Vatican City, or New Delhi <
Object level security – secure table and column names in addition to the data within them
Enhanced connectors – support for Tabular Object Model APIs and TMSL scripting
Graph data support: SQL Server offers graph database capabilities to model many-to-many relationships. The graph relationships are integrated into Transact-SQL and receive the benefits of using SQL Server as the foundational database management system. A graph database is a collection of nodes (or vertices) and edges (or relationships). A node represents an entity (for example, a person or an organization) and an edge represents a relationship between the two nodes that it connects (for example, likes or friends). Both nodes and edges may have properties associated with them.
<
GPU – offload some of heavy compute to the graphics processing unit (GPU). As a result, applications simply run faster. This is particularly applicable to deep learning, analytics, and engineering applications.
Native scoring with T-SQL - Microsoft created an extensibility framework that allows R scripts to be executed from T-SQL. This framework supports any operation you might perform in R, ranging from simple functions to training complex machine learning models. However, the dual-process architecture requires invoking an external R process for every call, regardless of the complexity of the operation. If you are loading a pre-trained model from a table and scoring against it on data already in SQL Server, the overhead of calling the external R process represents an unnecessary performance cost. Scoring is a two-step process. First, you specify a pre-trained model to load from a table. Second, pass new input data to the function, to generate prediction values (or scores). The input can be either tabular or single rows. You can choose to output a single column value representing a probability, or you might output several values, such as a confidence interval, error, or other useful complement to the prediction. <
*Available for Windows Server only

18. System Center Marketing
8/22/2018
SQL SERVER 2017 ENTERPRISE
End-to-end mobile BI on all major platforms
Enhanced connectors, new transformations, object-level security, ragged hierarchies**
Graph data support
Access to Power BI Report Server with SA
Enhanced direct query
In-memory analytics
Advanced data mining
Advanced tabular models
Web portal experience (all reports in one place)
Modernized reports
Pin report to Power BI
Enhanced multi- dimensional models
Basic Adaptive Query Processing (Interleaved execution)
Advanced Adaptive Query Processing (Batch mode memory grant feedback, Batch mode adaptive joins)
Enhanced in-memory ColumnStore
PolyBase in scale-out configuration (head and compute nodes)
Deployment rights for APS
Distributed query processing
Support for JSON
Always Encrypted
Row-level security
Dynamic data masking
Enhanced separation of duties
Enhanced SQL Server auditing
Transparent data encryption
OS max cores and memory
Enhanced in-memory OLTP performance
Operational analytics
Enhanced AlwaysOn with no domain join (WS 2016)
Query Store
Temporal
Automatic Plan Correction
Python integration - ability to run on GPUs and full parallelism through Machine Learning Services**
Machine Learning for Hadoop/Spark and Machine Learning for Linux, a software assurance benefit
In database Advanced Analytics
R integration with massive parallel processing for performance and scale
Works with in- memory technology
Run in database or standalone
Connectivity to R Open
Data Warehousing
OLTP Performance
Security
Business Intelligence
Advanced Analytics
Stretch Database
Enhanced backup to Azure
Enhanced HA and DR with Azure – ease of use, no domain join (WS2016)
SSIS integration with Azure Data Factory and Azure SQL Data Warehouse
Hybrid Cloud
Linux support
Container support
Platform
Enterprise
Slides features only the core new capabilities in Enterprise (when a comparison with Standard is not needed).
*Enterprise includes all Standard features. **Available for Windows Server only.
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. 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 provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

19. System Center Marketing
8/22/2018
SQL SERVER 2017 STANDARD
Enhanced connectors, new transformations, object-level security, ragged hierarchies**
Graph data support
Basic tabular (16GB memory per instance)
Modernized reports
Pin report to Power BI
Enhanced multi- dimensional models
Basic Adaptive Query Processing (Interleaved execution)
PolyBase (compute node only)
Support for JSON
Database snapshot
Partitioning
Compression
In-memory ColumnStore*
Change data capture
Always Encrypted
Row-level security
Dynamic data masking
Basic auditing
Separation of duties
In-memory OLTP (subject to memory limits)
Operational analytics
24 cores max and 128 GB max memory
2-node single database failover (non-readable secondary)
Query Store
Temporal
Automatic Plan Correction
Single-threaded for RRE
Connectivity to R Open and Python, Limited parallelism**
Data Warehousing
OLTP Performance
Security
Business Intelligence
Advanced Analytics
Standard
Stretch Database
Enhanced backup to Azure
Hybrid Cloud
Linux support
Container support
Platform
Slides features only the core new capabilities in Standard (when a comparison with Enterprise is not needed).
*In-memory columnstore and Operational Analytics limited to 32 GB of memory and 2 parallel cores. **Available for Windows Server only.
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. 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 provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

20. WHAT’S NEW IN SQL SERVER 2017 SINCE 2014
System Center Marketing
8/22/2018
WHAT’S NEW IN SQL SERVER 2017 SINCE 2014
OLTP Performance
Security
Data Warehousing
Business Intelligence
Advanced Analytics
Hybrid Cloud
Real-time operational analytics with in-memory OLTP or on disk
In-memory for more applications
Unparalleled scalability with Windows Server 2016, with 12TB memory and Windows Server max cores
Multiple node failover clustering (3 synchronous, up to 8 replicas)
SQL Server Development Tools in Visual Studio
Query Store
Temporal support
Automatic Plan Correction
Transparent Data Encryption
Always Encrypted
Row-level security
Dynamic data masking
Adaptive Query Processing
Operational analytics
In-memory ColumnStore
Deployment rights for APS
Enhanced In-memory ColumnStore for DW
PolyBase for simple T-SQL to query structured and unstructured data
Enhanced database caching
Up to 15,000 partitions
Enhanced connectors, new transformations, object-level security, ragged hierarchies**
Graph data support
Mobile BI
Enhanced SSIS
Enterprise-grade Analysis Services
Advanced tabular models
JSON support, enhanced DQS
Enhanced MDS
Enhanced Reporting Services
Temporal tables
In-memory analytics
Advanced data mining
Access to reports online or offline
Create mobile reports using the SQL Server Mobile Report Publisher
Consume with Power BI mobile apps
Python integration**
R built-in to your T-SQL
RRE APIs with full parallelism and no memory limits for scale/performance
Built-in In-memory Advanced Analytics
Advanced tabular model
Direct query
Advanced data mining
Stretch database
Partitioning for efficient data loading
Hybrid scenarios with SSIS
Enhanced backup to Azure
Easy migration to the cloud
Simplified cloud DR with AlwaysOn replicas
Complete inventory of features since 2008 R2 by pillar for customers upgrading from SQL Server 2014 (heroes features on earlier slides).
Platform
Linux support
Container support
**Available for Windows Server only.
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. 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 provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

21. WHAT’S NEW IN SQL SERVER 2017 SINCE 2012
System Center Marketing
8/22/2018
WHAT’S NEW IN SQL SERVER 2017 SINCE 2012
OLTP Performance
Security
Data Warehousing
Business Intelligence
Advanced Analytics
Hybrid Cloud
Real-time operational analytics with in-memory OLTP or on disk
In-memory for more applications
Unparalleled scalability with Windows Server 2016, with 12TB memory and Windows Server max cores
Enhanced AlwaysOn, with 8 secondaries and Replica Wizard
Multiple node failover clustering (3 synchronous, up to 8 replicas)
In memory OLTP
Buffer Pool Extension to SSDs
Enhanced query processing
Resource Governor adds IO governance
SysPrep as cluster level
Predictable performance with tiering of compute, network and story with Windows Server 2012 R2
Delayed Durability
Clustered Shared Volume support, VHDX support (Windows Server R2)
Manage on-premises and cloud apps (System Center 2012 R2)
Query optimization enhancements
Query Store
Temporal support
Automatic Plan Correction
Transparent Data Encryption
Always Encrypted
Row-level security
Dynamic data masking
Enhanced separation of duties
CC certification at High Assurance Level for 2014
Backup encryption support
Adaptive Query Processing
Operational analytics
In-memory ColumnStore
Deployment rights for APS
Enhanced In-memory ColumnStore for DW
PolyBase for simple T-SQL to query structured and unstructured data
Enhanced database caching
Up to 15,000 partitions
Analytics Platform System
Enhanced connectors, new transformations, object-level security, ragged hierarchies**
Graph data support
Mobile BI
Enhanced SSIS
Enterprise-grade Analysis Services
Advanced tabular models
Enhanced multidimensional models
JSON support
Enhanced DQS, enhanced MDS
Enhanced Reporting Services
Temporal tables
In-memory analytics
Advanced data mining
Create mobile reports using the SQL Server Mobile Report Publisher
Consume with Power BI mobile apps on all major platforms
Azure HDInsight Service
Pin report items to Power BI
Power Map for Excel
Mobile BI interfaces for Power BI
Python integration**
R built-in to your T-SQL
RRE APIs with full parallelism and no memory limits for scale/performance
Built-in In-memory Advanced Analytics
Advanced tabular model
Direct query
Advanced data mining
Stretch database
Partitioning for efficient data loading
Hybrid scenarios with SSIS
Enhanced backup to Azure
Easy migration to the cloud
Simplified cloud DR with AlwaysOn replicas
Simplified backup to Azure
Support for backup of previous versions of SQL Server to Azure
Cloud back-up encryption support
Simplified cloud Disaster Recovery with AlwaysOn replicas in Azure VMs
New Azure Deployment UI for SQL Server
Larger SQL Server VMs and memory sizes available in Azure
Complete inventory of features since 2012 by pillar for customers upgrading from SQL Server 2012.
Platform
Linux support
Container support
**Available for Windows Server only.
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. 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 provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

22. WHAT’S NEW IN SQL SERVER 2017 SINCE 2008 R2
System Center Marketing
8/22/2018
WHAT’S NEW IN SQL SERVER SINCE 2008 R2
OLTP Performance
Security
Business Intelligence
Hybrid Cloud
Real-time operational analytics with in-memory OLTP or on disk
In-memory for more applications
Unparalleled scalability with Windows Server 2016, with 12TB memory and Windows Server 2016 max cores
Enhanced AlwaysOn, with 8 secondaries and Replica Wizard
Multiple node failover clustering (3 synchronous, up to 8 replicas)
In memory OLTP
Buffer Pool Extension to SSDs
Enhanced query processing
Resource Governor adds IO governance
SysPrep as cluster level
Predictable performance with tiering of compute, network and story with Windows Server 2012 R2
Delayed Durability
Clustered Shared Volume support, VHDX support (Windows Server 2012 R2)
Manage on-premises and cloud apps (System Center 2012 R2)
Query optimization enhancements
Recovery Advisor
Windows Server Core
Live Migration
Online operations enhancements
Query Store
Temporal support
SQL Server Data Tools
Local DB runtime (Express)
Data-tier application component project template
Data-Tier Application Framework (DAC Fx)
Interoperability support (ADO.NET, ODBC, JDBC, PDO, ADO APIs and .NET C/C++, Java, Linux and PHP platforms)
Enhanced support for ANSI SQL standards
Transact-SQL Static Code Analysis tools
Transact-SQL code snippets
Intellisense
FileTable build on FILESTREAM
Remote Blob Storage with SharePoint 2010
Statistical Semantic Search
Spatial features, Full Globe and arcs
Large user-defined data types
Distributed Replay
Contained Database Authentication
System Center Management Pack for SQL Server 2012
Windows PowerShell 2.0 support
Multi-server Management with SQL Server Utility Control Point
Data Tier Application Component
Automatic Plan Correction
Transparent Data Encryption
Always Encrypted
Enhanced separation of duty
Row-level security
Dynamic data masking
Enhanced separation of duties
Default schema for groups
SQL Server Audit
SQL Server fine-grained auditing
Enhanced connectors, new transformations, object-level security, ragged hierarchies**
Graph data support
Mobile BI
Enhanced SSIS
Enterprise-grade Analysis Services
Advanced tabular models
In-memory analytics
Enhanced multidimensional models
JSON support
Enhanced DQS
Enhanced MDS
Modern Reporting Services
Temporal tables
Advanced data mining
Create mobile reports using the SQL Server Mobile Report Publisher
Consume with Power BI mobile apps
Azure HDInsight Service
Power BI
Power Map for Excel
Mash up data from different sources, such as Oracle & Hadoop
HA for StreamInsight, complex event processing
SQL Server Data Tools support for BI
Change Data Capture for Oracle
Import PowerPivot models into Analysis Services
Enhanced productivity and performance
Power View
Configurable reporting alerts
Reporting as SharePoint Shared Service
Build organization knowledge base
Connect to 3rd party data cleansing providers
Master Data Hub
Master Data Services Add-in for Excel
Graphical tools in SSIS
Extensible object model
SSIS as a Server
Broader data integration with more sources: DB vendors, cloud, Hadoop
Pipeline improvements
Stretch database
Partitioning for efficient data loading
Hybrid scenarios with SSIS
Enhanced backup to Azure
Easy migration to the cloud
Simplified cloud DR with AlwaysOn replicas
Simplified backup to Azure
Support for backup of previous versions of SQL Server to Azure
Cloud back-up encryption support
Simplified cloud Disaster Recovery with AlwaysOn replicas in Azure VMs
New Azure Deployment UI for SQL Server
Larger SQL Server VMs and memory sizes available in Azure
SQL Server Data Tools
Snapshot backups to Azure via SQL Server Management Studio
Data Warehousing
Complete inventory of features since 2008 R2 by pillar for customers upgrading from SQL Server 2008 R2.
Advanced Analytics
Adaptive Query Processing
Operational analytics
In-memory ColumnStore
Deployment rights for APS
Enhanced In-memory ColumnStore for DW
PolyBase for simple T-SQL to query structured and unstructured data
Enhanced database caching
Up to 15,000 partitions
Analytics Platform System
Python integration**
R built-in to your T-SQL
RRE APIs with full parallelism and no memory limits for scale/performance
Built-in In-memory Advanced Analytics
Advanced tabular model
Direct query
Advanced data mining
SSDT in Visual Studio
Platform
Linux support
Container support
**Available for Windows Server only.
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. 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 provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

23. The modern data estate SQL Server R Microsoft’s on-premises solution
1/10th the cost of Oracle
Operational databases
Industry leader 4 years in a row 
Data warehouses
#1 TPC-H performance
Data lakes
T-SQL query over any data
SQL Server is Microsoft’s on-premises solution for the Modern Data Estate, with everything-built in.
Reason over any data, anywhere
Flexibility of choice
Security and privacy
Only commercial DB with AI built-in R
Any language, any platform, anywhere
MULTI
JAVA
SQL Server
Most secure over last 7 years

24. Revert to previously effective plan
Automatically fix problems without tuning Better performance with Automatic Plan Correction
Continuous performance plan monitoring and analysis
Detect problematic plans
Automatically fix performance problems caused by SQL plan choice regressions
Query times
Plan 1
Plan 2
Plan 3
Plan 2
Revert to previously effective plan

25. Adaptive Query Processing feature family
Interleaved Execution
Batch Mode Memory Grant Feedback
Batch Mode Adaptive Joins

26. Adaptive Query Processing Interleaved Execution
Pre 2017
Optimize
Execute
100 rows guessed for MSTVFs
Performance issues if skewed
Problem
Multi-statement table valued functions (MSTVFs) are treated as a black box by QP and a fixed optimization guess is used
Solution
Interleaved Execution materializes rows and gathers row counts for MSTVFs
Downstream operations benefit from the corrected MSTVF cardinality estimate
2017+
MSTVF
identified
500k rows
assumed
Execute
Good
Performance
Optimize

27. Adaptive Query Processing Batch Mode Memory Grant Feedback
Problem
Queries may spill to disk or take too much memory based on poor cardinality estimates
Solution
Memory Grant Feedback adjusts memory grants based on execution feedback
Memory Grant Feedback removes spills and improves concurrency for repeating queries
Before
After
101010
Query
Adaptive Query Processing
Learn
Memory grant
Run query
Spill to disk
All in memory

28. Adaptive Query Processing Batch Mode Adaptive Joins
Problem
If cardinality estimates are skewed, an inappropriate join algorithm may be chosen
Solution
Adaptive Joins defers the choice of hash join or nested loop until after the first join input has been scanned
Adaptive Joins uses nested loop for small inputs, hash joins for large inputs
Build Input
Hash join
Yes
Adaptive Threshold
Nested loop No

29. SQL Server Python SQL Server 2016 introduced SQL Server R Services
SQL Server 2017 now introduces SQL Server Machine Learning Services which includes Python and R
Two options to execute inside SQL Server:
Start from Python and execute remote jobs
Start from T-SQL with an embedded Python script
Secure and fully governed feature
Supports SQL and integrated authentication
Manage resources (CPU & memory etc.)
Source:

30. Native scoring
New PREDICT function in Transact-SQL to perform scoring even if R isn't installed
Train the model using one of the supported RevoScaleR and RevoScalePy algorithms
Save the model
Perform low-latency scoring
varbinary(max) = “SELECT TOP from [models_table]”;
= “SELECT ID, [Gender], [Income] from NewCustomers”;
SELECT PREDICT [class]
FROM PREDICT( MODEL DATA
WITH (class string);

31. Bachelor of Science, Finance
generate insights across diverse data New relationships uncovered with graph data support
Professional networking app
Bachelor of Science, Finance
Statistics
Skill
Degree earned
Bring graph data support to your relational data store
Analyze interconnected data and generate deeper insights
Query data stored in Hadoop with PolyBase
Get value and insight from data lakes using Hadoop combined with SQL Server
Contoso
Andy Smith
Skill
Former employer
Position
Coworker
Mary Jones
Business analyst
Role in company
Employer
The power to query over any type of data with New support for Graph Data
But, when it comes to hierarchical data with complex relationships or data that share multiple relationships, one might find themselves struggling with a good schema design to represent all the entities and relationships, and writing optimal queries to analyze complex data and relationships between the tables.
We live in an era of big data and connected information; people, machines, devices, businesses across the continents are connected to each other more than ever before. Analyzing connected information is becoming critical for businesses to achieve operational agility. Users are finding it easier to model data and complex relationships with the help of graph databases. Native graph databases have risen in popularity, being used for social networks, transportation networks, logistics, and much more. Graph database scenarios can easily be found across several business disciplines, including supply chain management, computer or telecommunication networks, detecting fraud attacks, and recommendation engines.
At Microsoft, we believe that there should be no need for our customers to turn to a new system just to meet their new or evolving graph database requirements. SQL Server is already trusted by millions of customers for mission-critical workloads and with graph extensions in SQL Server 2017, customers get best of both relational and graph databases in a single product, including the ability to query across all data using a single platform. Users can also benefit from other cutting-edge technologies already available in SQL Server, such as Columnstore indexes, Advanced Analytics using SQL Server R Services, High Availability and more.
When you manage Graph Data in SQL Server, you get:
Full CRUD support to create nodes and edges
T-SQL Query language extension to better support graph analysis, including multi-hop navigation using join-free pattern matching
SQL engine integration enables querying across SQL tables and graph data
Existing tools work out of the box with graph data
And of course, our discussion of SQL Server working with all the data in your data estate wouldn’t be complete without discussion of PolyBase.
You can create an external table that maps the two structured and unstructured data and the PolyBase technology available in SQL Server allows customers to query that external table, so the structured and unstructured data can be correlated together.
Employer
Position
Role in company
Program manager
AdventureWorks
Role in company