The SQL Server 2022 Workshop.pptx

Editor's Notes

• #6: Editions, licensing, and pricing come at GA
• #8: 2 mins
• #9: TODO: Get this slide back TODO: Update with new benchmarks
• #11: We will support restore from sql 2005 but oldest dbcompat is 100 (sql 2008) supported
• #13: Here is an article https://docs.microsoft.com/en-us/azure/azure-arc/servers/manage-vm-extensions Here is another one https://docs.microsoft.com/en-us/azure/azure-arc/servers/agent-overview#agent-component-details If you supply MS login it auto fills out required fields. You may want to change RG .It even provides a drop down to choose the RG per your sub. It failed though with this cryptic error probably because my login doesn’t have the right permissions.
• #15: RHEL container images are listed at https://mcr.microsoft.com/v2/mssql/rhel/server/tags/list Ubuntu images are listed at https://mcr.microsoft.com/v2/mssql/server/tags/list There is some new DTC functionality but I’m trying to determine if this is new to SQL22
• #17: Answers E And C
• #24: Assume a deployed SQL Server 2022 instance with a database and Azure SQL Managed Instance connected with Azure (VPN, Express Route, Azure virtual network, …) Establish connectivity between SQL Server and Azure SQL Managed Instance using Azure networking and dbm endpoints Start the replication database process on SQL Server 2022 by creating an AG (you could have an existing AG) on SQL Server 2022 with dbm endpoints secured on each side. Create the DAG with SQL Server 2022 and Azure SQL Managed Instance. Create the link between SQL Server and MI which seeds the database to MI Changes on the primary are sent to the replica database in MI via transaction log blocks Execute read queries against replicas in MI
• #29: Same concept as CosmosDB link but we use SQL pools so this is a warehouse solution with full scalability.
• #31: Powershell also supported to create the linked services and manage linked connection Guide to the numbers and animation: Start with SQL Serer 2022 deployed with a database and an Azure subscription Create a Synapse workspace Create a SQL dedicated pool Create a linked service for SQL Server 2022. Install SHIR to connect to the Synapse control plane Create an Azure Storage Account called a Landing Zone Create a linked service for the landing zone Create a linked connection for the linked service for SQL. Choose your tables. Start the connection. SHIR executes stored procedures to SQL Server 2022. Snapshots of tables are captured and submitted to the landing zone in the form of parquet and schema files. The Ingestion service in the control plane takes the landing zone files and creates tables in the SQL pools. Data is also inserted into the pool tables based on the initial snapshot. Any modifications to tables are recorded in the transaction log. Change feeds within SQL Server harvest log changes and put them in queue. Internal tasks publish changes to the landing zone in the form of CSV and manifest files. The ingestion service takes the landing zone files and send modifications to affected tables in the SQL pools. Create a Power I report to visualize your data directly on SQL pool tables. Details of how ADF talks to SHIR can be found at https://docs.microsoft.com/en-us/azure/data-factory/create-self-hosted-integration-runtime?tabs=data-factory#command-flow-and-data-flow
• #33: 2 mins
• #35: TODO: how does Arc extension help Defender like it does with AAD and Purview
• #36: You can configure via portal, powershell, or az cli
• #40: TODO: Checking on CLI Register SQL Server as an Arc-enabled SQL Server and enable AAD authentication. Then enable Microsoft Purview which writes information into the registry or mssql.conf Register SQL Server in Purview Studio and create Data and/or DevOps policies SQL Server uses information in the registry to extract policy information from Purview and store this in a policy cache in system table metadata A login using AAD tries to connect. SQL Server will use AAD auth and look at the policy cache to determine if the AAD account can connect via policies and applies permissions based on the policy.
• #43: Answers E And B
• #44: Answers True And B
• #70: Answers D And B
• #71: Answers E And B
• #75: Strict connected encryption can be forced on server with Network Config
• #79: Ledger limits In-memory tables aren't supported. Sparse column sets aren't supported. SWITCH IN/OUT partition isn't supported. Ledger tables can't have full-text indexes. Ledger tables can't be graph table. Ledger tables can't be FileTables. Ledger tables can't have a rowstore non-clustered index when they have a clustered columnstore index. Change tracking isn't allowed on the history table but is allowed on ledger tables. Change data capture isn't supported for ledger tables. Transactional replication isn't supported for ledger tables.
• #80: TODO: Review with latest SQL 2022 deck
• #86: 2 mins Hyperscale uses snapshot backups with Azure Standard Storage
• #87: There are several current customer demands that we have captured from customer and support team discussions including MVPs and architects in the field over the years. Stability for critical workloads – Customers want to reserve capacity for business processes and offload maintenance activities. For orders and operational processes to complete as expected, organizations must make sure they have the performance available for new orders and can continue their business continuity and disaster recovery (BCDR) tasks as well. Some customers find that the size of their databases and the overhead that compression causes, forces them to choose between maintenance activities and revenue. Consider the Black Friday / holiday windows, end of month & end of quarter reporting, and other spikes throughout the year.(Speaker notes: You could mention Mother’s Day & Valentines Day for a florist, tax season, Superbowl advertising, and other critical windows that need special considerations.) Consistent performance without oversizing – Customers need to minimize CPU spikes. The ability to flatten the resource consumption curve, for higher compute scenarios, provides predictable performance to end users. The goal is for customers to maintain peak workloads while allowing compute to normalize. Remove scalability blockers – Often customers find that they must over purchase hardware, and this is especially evident with virtual machines, where customers have to scale up the compute in order to handle high CPU tasks such as compression and encryption. Being able to move the high compute workloads away from the host and virtual machines would allow customers to right-size their solutions and maximize their total cost of ownership. Make the most of current hardware – Customers are interested in taking advantage of their hardware investments and extending the lifecycle of their current platforms. In SQL Server 2022, we have made investments in the SQL Server engine to provide a method of offloading the types of workloads described previously and in particular, high compute scenarios, to specialized hardware. We have also architected a foundational framework to provide accelerators for SQL Server workloads to extend the SQL Server’s engines ability to handle high compute tasks through offloading. <CLICK> The first implementation of this technology is through Microsoft’s partnership with Intel providing a new backup compression option that relies on hardware acceleration. In SQL Server 2022, we will support the Intel Quick Assist Technology (QAT) providing backup compression and hardware acceleration for SQL Server 2022. Customers will gain better compression with reduced capacity and will also see much improved performance. Better performance means faster backups and even more importantly, faster restores – with faster backups we can reduce our backup windows. And with faster restores we improve recovery SLAs. There are a number of existing backup compression products already on the market for SQL Server, but these are all software-based which adds overhead, and the backup speeds are unpredictable. Additionally, none of the existing software-based solutions are integrated with the SQL Server engine. With Intel Quick Assist Technology (QAT) we will provide several commonly used algorithms and formats, such as Gzip, Express, deflate, mszip, and zlib. The Intel QAT backup compression tightly integrates with SQL Server through T-SQL and customers will be able to choose when they want to use QAT backup compression or when they may want to use the Microsoft default by simply changing the algorithm command in T-SQL. During testing of Intel Quick Assist Technology (QAT) with SQL Server 2022 - we witnessed over a 2X speed improvement and a 6% reduction in the storage capacity consumed for the tested backups on Icelake processors and even better results on Sapphire Rapid architecture. With Intel Quick Assist Technology (QAT) and SQL Server 2022 – when customers need to backup their databases with compression, they will not see their backups struggle even when the server is under peak load and will gain better compression rates.
• #91: Answers C And B
• #92: Answers D And D
• #99: CETAS = CREATE EXTERNAL TABLE AS SELECT
• #100: 2 mins
• #101: 2 mins
• #104: Answers D And A
• #109: Answers D And A
• #116: Get details for SSAS at https://powerbi.microsoft.com/en-us/blog/whats-new-in-sql-server-2022-analysis-services-ctp-2-0 Get details for SSRS at https://powerbi.microsoft.com/en-us/blog/reminder-of-features-being-removed-with-the-next-release-of-sql-server/ TODO: Still waiting for Cookie McCray for details on these