24 October 2017

SQL Server 2017 on Linux



SQL Server 2017 on Linux

Microsoft has heard from you, our customers, that your data estate gets bigger, more complicated, and more diverse every year. You need solutions that work across platforms, whether on-premises or in the cloud, and that meet your data workloads where they are. Embracing this choice, earlier today we announced the general availability of SQL Server 2017 on Linux, Windows, and Docker on October 2, 2017.



Today, Microsoft and Red Hat are delivering on choice by announcing the availability of Microsoft SQL Server 2017 on Red Hat Enterprise Linux, the world’s leading enterprise Linux platform. As Microsoft’s reference Linux platform for SQL Server, Red Hat Enterprise Linux extends the enterprise database and analytics capabilities of SQL Server by delivering it on the industry-leading platform for performance, security features, stability, reliability, and manageability.

Customers will be able to bring the performance and security features of SQL Server to Linux workloads. SQL Server 2017 on Red Hat Enterprise Linux delivers mission-critical OLTP database capabilities and enterprise data warehousing with in-memory technology across workloads. SQL Server 2017 embraces developers by delivering choice in language and platform, with container support that seamlessly facilitates DevOps scenarios. The new release of SQL Server delivers all of this, built-in. And, it runs wherever you want, whether in your datacenter, in Azure virtual machines, or in containers running on Red Hat OpenShift Container Platform!



Also starting October 2nd until June 30th, 2018, we are launching a SQL Server on Red Hat Enterprise Linux offer to help with upgrades and migrations. This offer provides up to 30% off SQL Server 2017 through an annual subscription. When customers purchase a new Red Hat Enterprise Linux subscription to support their SQL Server, they will be eligible for another 30% off their Red Hat Enterprise Linux subscription price.

In addition to discounts on SQL Server and Red Hat Enterprise Linux, all of this is backed by integrated support from Microsoft and Red Hat.

Bootcamp 2017 - SQL Server on Linux


 SQL Server 2017 is generally available for purchase and download! The new release is available right now for evaluation or purchase through the Microsoft Store, and will be available to Volume Licensing customers later today. Customers now have the flexibility for the first time ever to run industry-leading SQL Server on their choice of Linux, Docker Enterprise Edition-certified containers and, of course, Windows Server. It’s a stride forward for our modern and hybrid data platform across on-premises and cloud.

Everything you need to know about SQL Server 2017


In the 18 months since announcing our intent to bring SQL Server to Linux, we’ve been focused on making SQL Server perform and scale to the industry-leading levels customers expect from SQL Server, making SQL Server feel familiar yet native to Linux, and ensuring compatibility between SQL Server on Windows and Linux. With all the enterprise database features you rely on, from Active Directory authentication, to encryption, to Always On availability groups, to record-breaking performance, SQL Server is at parity on Windows and Linux. We have also brought SQL Server Integration Services to Linux so that you can perform data integration just like on Windows. SQL Server 2017 supports Red Hat Enterprise Linux, SUSE Linux Enterprise Server, and Ubuntu.

There are a number of new features for SQL Server that we think make this the best release ever. Here are just a few:

  • Container support seamlessly facilitates your development and DevOps scenarios by enabling you to quickly spin up SQL Server containers and get rid of them when you are finished. SQL Server supports Docker Enterprise Edition, Kubernetes and OpenShift container platforms.
  • AI with R and Python analytics enables you to build intelligent apps using scalable, GPU-accelerated, parallelized R and now Python analytics running in the database.
  • Graph data analysis will enable customers to use graph data storage and query language extensions for graph-native query syntax in order to discover new kinds of relationships in highly interconnected data.
  • Adaptive Query Processing is a new family of features in SQL Server that bring intelligence to database performance. For example, Adaptive Memory Grants in SQL Server track and learn from how much memory is used by a given query to right-size memory grants.
  • Automatic Plan Correction ensures continuous performance by finding and fixing performance regressions.



Above and beyond these top-line features, there are more enhancements that you haven’t heard as much about, but we hope will truly delight you:


  • Resumable online index rebuild lets you stop and start index maintenance. This gives you the ability to optimize index performance by re-indexing more frequently – without having to wait for a long maintenance window. It also means you can pick up right where you left off in the event of a disruption to database service.
  • LOB compression in columnstore indexes. Previously, it was difficult to include data which contained LOBs in a columnstore index due to size. Now those LOBs can be compressed, making LOBs easier to work with and broadening the applicability of the columnstore feature.
  • Clusterless availability groups enable you to scale out reads by building an Always On availability group without having to use an underlying cluster.
  • Continued improvement to key performance features such as columnstore, in-memory OLTP, and the query optimizer to drive new record-setting performance. We’ll share some even more exciting perf and scale numbers soon!
  • Native scoring in T-SQL lets you score operational data using advanced analytics in near real-time because you don’t have to load the Machine Learning libraries to access your model.
  • SQL Server Integration Services (SSIS) scale-out enables you to speed package execution performance by distributing execution to multiple machines. These packages are executed in parallel, in a scale-out mode.
What’s new in SQL Server 2017





Many enhancements were made to SQL Server Analysis Services including:
  • Modern “get data” experience with a number of new connectors like Oracle, MySQL, Sybase, Teradata, and more to come. New transformations enable mashing up of the data being ingested into tabular models.
  • Object-level security for tables and columns.
  • Detail rows and ragged hierarchies support, enabling additional drill-down capabilities for your tabular models.
Enhancements were made to SQL Server Reporting Services as well, including:
  • Lightweight installer with zero impact on your SQL Server databases or other SQL Server features.
  • REST API for programmatic access to reports, KPIs, data sources, and more.
  • Report comments, enabling users to engage in discussion about reports.



In addition to the ability to upgrade existing SQL Server to 2017, there are a few more benefits to renewing your software assurance:


  • Machine Learning Server for Hadoop, formerly R Server, brings R and Python based, scalable analytics to Hadoop and Spark environments, and it is now available to SQL Server Enterprise edition customers as a Software Assurance benefit.
  • SQL Server Enterprise Edition Software Assurance benefits also enable you to run Power BI Report Server. Power BI Report Server enables self-service BI and enterprise reporting, all in one solution by allowing you to manage your SQL Server Reporting Services (SSRS) reports alongside your Power BI reports. Power BI Report Server is also included with the purchase of Power BI Premium.
  • Lastly, but importantly, we are also modernizing how we service SQL Server. Please see our release management blog for all the details on what to expect for servicing SQL Server 2017 and beyond.

Microsoft will continue to invest in SQL Server 2017 and cloud-first development model, to ensure that the pace of innovation stays fast.

SQL Server 2017 sets the standard when it comes to speed and performance. Based on the incredible work of SQL Server 2016 (See the blog series It Just Runs Faster), SQL Server 2017 is fast: built-in, simple, and online. Maybe you caught my presentation at Microsoft Ignite where I demonstrated 1 million transactions per minute on my laptop using the popular tool HammerDB¹ by simply installing SQL Server out of the box with no configuration changes (with the HammerDB client and SQL Server on the same machine!)

SQL Server 2017 on Linux Introduction


Consider for a minute all the built-in capabilities that power the speed of SQL Server. From a SQLOS scheduling engine that minimizes OS context switches to read-ahead scanning to automatic scaling as you add NUMA and CPUs. And we parallelize everything! From queries to indexes to statistics to backups to recovery to background threads like LogWriter. We partition and parallelize our engine to scale from your laptop to the biggest servers in the world.

Like the enhancements we made as described in It Just Runs Faster, in SQL Server 2016, we are always looking to tune our engine for speed, all based on customer experiences. Take, for example, indirect checkpoint, which is designed to provide a more predictable recovery time for a database. We boosted scalability of this feature based on customer feedback. We also made scalability improvements for parallel scanning and consistency check performance. No knobs required. Just built-in for speed.

One of the coolest performance aspects to built-in speed is online operations. We know you need to perform other maintenance tasks than just run queries, but keep your application up and running, so we support online backups, consistency checks, and index rebuilds. SQL Server 2017 enhances this functionality with resumable online index builds allowing you to pause an index build and resume it at any time (even after a failure).

Microsoft SQL Server 2017 Deep Dive


SQL Server 2017 is faster than you think. SQL Server 2017 was designed from the beginning to run fast on popular Linux distributions such as Red Hat Enterprise Linux, SUSE Linux Enterprise, and Ubuntu whether that is on your server or a Docker Container. Don’t believe it? Check out our world record 1TB TPC-H benchmark result (non-clustered) for SQL Server on Red Hat Enterprise Linux. Even though this is our first release on Linux, we know how to configure and run SQL Server on Linux for maximum speed. Read our best practices guide for performance settings on Linux in our documentation. We know it performs well because our customers tell us. Read the amazing story of dv01 and how SQL Server on Linux exceeded their performance expectations as they migrated from PostgreSQL

SQL Server 2017 Deep Dive - @Ignite 2017


One of the key technologies to achieve a result like this is columnstore indexes. This is one of the most powerful features of SQL Server for high-speed analytic queries and large databases. Columnstore indexes boost performance by organizing columnar data in a new way than traditional indexes, compressing data to reduce memory and disk footprint, filtering scans automatically through rowgroup elimination and processing queries in batches. SQL Server runs at warp speed for data warehouses and columnstore is the fuel. At Microsoft Ignite, I demonstrated how columnstore indexes can make PowerBI with Direct Query against SQL Server faster handling the self-service, ad-hoc nature of PowerBI queries.

Microsoft Ignite 2017 - SQL Server on Kubernetes, Swarm, and Open Shift


SQL Server also excels at transaction processing, the heart of many top enterprise workloads. Got RAM? Not only does columnstore use in-memory technologies to achieve speed, but our In-Memory OLTP feature focuses on optimized access to memory-optimized tables. This feature is named OLTP, but it can be so much more. ETL staging tables, IoT workloads, table types (no more tempdb!), and “caching” tables. One of our customers was able to get a throughput of 1.2M batch requests/sec using SCHEMA_ONLY memory-optimized tables. To really boost transaction processing, also consider using SQL Server’s support for Persistent Memory (NVDIMM-N) and our optimization for transaction log (get ready for WRITELOG waits = 0!) performance. SQL Server 2017 supports any Persistent Memory technology supported on Windows Server 2016 and later releases.

Many customers I talk to have great performance when they first deploy SQL Server and their application. Keeping SQL Server fast and tuned is more of the challenge. SQL Server 2017 comes with features to keep you fast and tuned automatically and adaptively. Our Query Processing engine has all types of capabilities to create and build query plans to maximize the performance of your queries. We have created a new feature family in SQL Server 2017 to make it smarter, called Adaptive Query Processing. Imagine running a query that is not quite the speed you expect because of insufficient memory grants (which is a thorn in the side of many SQL Server users, as it can lead to a spill to tempdb). With Adaptive Query Processing, future executions of this query will have a corrected calculated memory grant avoiding the spill, all without requiring a recompilation of the query plan. Adaptive Query Processing handles other scenarios such as adaptive joins and interleaved execution of Table Valued Functions.

Choosing technologies for a big data solution in the cloud


Another way to keep you tuned is the amazing feature we added in SQL Server 2016 called Query Store. Query Store provides built-in capabilities to track and analyze query performance all stored in your database. For SQL Server 2017, we made tuning adjustments to Query Store to make it more efficient based on learnings in our Azure SQL Database service where Query Store is enabled for millions of databases. We added wait statistics so now you have an end-to-end picture of query performance. Perhaps though the most compelling enhancement in SQL Server 2017 is Automatic Tuning. Parameter Sniffing got you down? Automatic Tuning uses Query Store to detect query plan regressions and automatically forces a previous plan that used to run fast. What I love about this feature is that even if you don’t have it turned on, you can see recommendations it has detected about plan regressions. Then you can either manually force plans that you feel have regressed or turn on the feature to have SQL Server do it for you.

Introduction to PolyBase


SQL Server 2017 is the fastest database everywhere you need it. Whether it is your laptop, in your private cloud, or in our Azure public cloud infrastructure. Whether it is running on Linux, Windows, or Docker Containers, we have the speed to power any workload your application needs.

As I mentioned above, back in April, we announced our world record TPC-H 1TB data warehousing workload (non-clustered) for SQL Server 2017 running on a HPE ProLiant DL380 Gen9 using RedHat Enterprise Linux².

Perhaps you missed the announcement in June of 2017, of a new world record TPC-E benchmark result³ on SQL Server 2017 on Windows Server 2016 running on a Lenovo ThinkSystem SR650 continuing to demonstrate our leadership in database performance. This benchmark running on a 2 socket system using Intel’s Xeon Scalable Processors has set a new standard for price and performance, becoming the first TPC-E benchmark result ever to be under $100/tpsE.

We continued to show our proven speed for analytics by announcing in July of 2017 a new TPC-H 10TB (non-clustered) world record benchmark result4 of 1,336,109 QppH on Windows Server 2016 using a Lenovo ThinkSystem SR950 system with 6TB RAM and 224 logical CPUs.

While benchmarks can show the true speed of SQL Server, we believe it can perform well with your workload and maximize the computing power of your server. Perhaps you caught the session at Ignite where my colleague Travis Wright showed how we can scan a 180 Billion row table (from a 30TB database) in our labs in under 20 seconds powering 480 CPUs to 100% capacity. And if you don’t believe SQL Server is deployed in some of the biggest installations and servers in the world, I recently polled some of our field engineers, SQL Customer Advisor Team, and MVPs asking them for their largest SQL Server deployments. Over 30 people responded, and the average footprint of these installations was 3TB+ RAM on machines with 128 physical cores. Keep in mind that SQL Server on can theoretically scale to 24TB RAM on Windows and 64TB on Linux. And it supports the maximum CPUs of those systems (64 sockets with unlimited cores on Windows and 5120 logical CPUs on Linux). Look for more practical and fun demonstrations of the speed of SQL Server in the future.

Microsoft cloud big data strategy


It could be that you are consolidating your deployments and want to run SQL Server using Azure Virtual Machine, but not sure if the capacity is there for your performance needs. Consider that Azure Virtual machine has the new M-Series, which supports up to 128 vCPUs, 2TB RAM, and 64 Data Disks with a capacity of 160,000 IOPS. It could be that in your environment you want to scale out your read workload with Availability Group secondary replicas but don’t want to invest in Failover Clustering. SQL Server 2017 introduces the capability of read-scale availability groups without clustering supported both on Windows and Linux. Two other very nice performance features new to SQL Server 2017 are SSIS Scale Out, for those with data loading needs, and native scoring, which integrates machine learning algorithms into the SQL Server engine for maximum performance.

Microsoft Technologies for Data Science 201612


SQL Server 2017 brings to the database market a unique set of features and speed. A database engine that is fast, built-in with the power to scale, and even faster when taking advantage of technologies like columnstore Indexes and In-Memory OLTP. An engine that provides automation and adapts to keep you fast and tuned. And the fastest database everywhere you need it.

Machine learning services with SQL Server 2017

More Information:

https://docs.microsoft.com/en-us/sql/sql-server/sql-server-2017-release-notes

http://www.databasejournal.com/features/mssql/slideshows/9-new-features-with-sql-server-2017.html

https://www.microsoft.com/en-us/sql-server/sql-server-2017

https://myignite.microsoft.com/sessions

https://blogs.technet.microsoft.com/dataplatforminsider/2017/10/02/sql-server-2017-on-windows-linux-and-docker-is-now-generally-available/

https://blogs.msdn.microsoft.com/bobsql/

http://www.hammerdb.com

http://www.hammerdb.com/benchmarks.html

https://blogs.msdn.microsoft.com/sqlserverstorageengine/

https://blogs.technet.microsoft.com/dataplatforminsider/2017/09/28/enhancing-query-performance-with-adaptive-query-processing-in-sql-server-2017/

https://blogs.technet.microsoft.com/dataplatforminsider/2017/09/29/view-on-demand-microsoft-data-platform-sql-server-2017-and-azure-data-services/

https://blogs.technet.microsoft.com/dataplatforminsider/2017/10/10/whats-new-in-sql-server-management-studio-17-3/

https://redmondmag.com/articles/2017/09/25/microsoft-launches-sql-server-2017.aspx

https://redmondmag.com/articles/2017/04/19/sql-server-2017-preview.aspx

https://arstechnica.com/gadgets/2017/09/microsoft-ignite-2017-azure-sql/











Share:

0 reacties:

Post a Comment