Storing bits and bytes in the cloud Keith Burns Data Platform Architect Microsoft UK keith.burns@microsoft.com

SQL Azure Database Sync Service Reporting Extending SQL Server to the Cloud Database Sync Service Reporting SQL Azure Database Familiar SQL Server relational database model Support for existing APIs & tools Built for the cloud with high availability & fault tolerance Easily provision and manage databases across multiple datacenters Data Sync Service (CTP) Provides two-way sync of SQL Azure Databases across datacenters Example service available in the SQL Azure Labs environment SQL Azure Reporting (CTP) Reporting Services running on the Windows Azure Platform

SQL Azure Database

SQL Azure database service SQL Azure provides logical SQL Server Looks like SQL Server to TDS Client Actual data stored on multiple backend data nodes Logical optimisations supported Indexes, Query plans etc.. Physical optimisations not supported File Groups, Partitions etc… Transparently manages physical storage

Working with SQL Azure Databases Your App SQL Azure TDS Change Connection String Primary SQL Azure database Secondary database copies

Behind the Scenes of SQL Azure Apps use standard SQL client libraries: ODBC, ADO.Net, PHP, … Application Internet Load balancer forwards ‘sticky’ sessions to TDS protocol tier TDS (tcp) LB Security Boundary TDS (tcp) Gateway Gateway: TDS protocol gateway, enforces AUTHN/AUTHZ policy; proxy to backend SQL TDS (tcp) SQL SQL SQL SQL SQL SQL Scalability and Availability: Fabric, Failover, Replication, and Load balancing

Application Topologies From Windows Azure From Outside Microsoft Datacenter From Windows Azure & Outside Microsoft Datacenter SQL Server App Code / Tools Application / Browser App Code / Tools SQL Azure Data Sync Microsoft Datacenter Microsoft Datacenter Microsoft Datacenter Windows Azure Windows Azure SQL Azure SQL Azure SQL Azure Code Near Code Far Hybrid

SQL Azure Compatibility http://bit.ly/9fgzue Currently Supported Not Currently Supported Tables, indexes and views Data types XML, Hierarchyid, Spatial Stored Procedures Triggers Constraints Table variables, session temp tables (#t) Column & expression collation MARS Data Types Sparse Columns, Filestream Partitions Full-text indexes, indexed views SQL-CLR (MS use only) SQL Agent Global temp tables Server or database collations Distributed txns TDE, Compression, Auditing Tables require clustered indexes

Database Editions 1 GB 5 GB 10GB, 20GB, 30GB, 40GB, 50GB Web Edition Business Edition This is the maximum size we will not let you grow beyond You will only be charged for the actual peak size rounded up in any one day For example, a 3.4 GB Web Edition will be charged 5GB rate. CREATE DATABASE foo (EDITION=‘web’, MAXSIZE=1GB); ALTER DATABASE foo MODIFY (EDITION=‘business', MAXSIZE=50GB);

Database Editions Subscription & Pay as you go pricing Web Edition: 1 GB @ £6.06/month (ie ~ £72 / year) 5 GB @ £30.28/month Business Edition: 10 GB /20GB / 30GB /40GB / 50GB @ £60.60 / 10GB / month Data Transfer Cost £0.06 / GB in and £0.09 / GB out NO cost for intra-datacentre data transfer Business Edition Up to 50 GB 10 GB increments Web Edition 1 GB or 5 GB

Demo

Few other things to consider Latency Connection throttling Reasons include Lock Consumption: Sessions using over 1million locks Excessive Log usage: Single txn consuming >1Gb txn Uncommitted Transactions: Single txn consuming > 20% of the log Excessive Tempdb usage: Session using > 5Gb of TempDB … range of error codes returned See http://bit.ly/cCWmEd

“Backup” Database Copy (Available NOW!) Copies asynchronously ie control returns before copy completes View copy status: “Select * from sys.dm_database_copies” http://bit.ly/cwquUe DACPAC (see http://www.sqlazurelabs.com) SQL Azure Restore (CTP Soon) Point in time database restore CREATE DATABASE database_destination_name AS COPY OF [source_server_name.]database_source_name DacImportExportCli.exe -s myserver.database.windows.net -d northwind -f C:\northwind.bacpac -x -u username -p password

SQL Azure Data Sync

SQL Azure Data Sync Scenario Overview One-Off txfr. Scenarios Tools Migration to/from SQL Server Transfer between SQL Azure servers/databases BCP SSMS Generate script wizard SQL Server Import & Export Data SSIS SQL Azure Database Copy SQL Azure Migration Wizard DAC DAC Logical Import/Export (see SQLAzureLabs) On-Going sync. Publish to cloud Share between on-prem and cloud apps Some SQL Azure scale-out cases Geo-locality One-way or two-way SQL Azure Data Sync Sync Framework

SQL Azure Data Sync Sync Powers movement of data Cloud  cloud On-premises  cloud Getting data where you need it Sync SQL Azure instances Sync SQL Server to SQL Azure Sync offline apps to SQL Azure Enable geo-replication of data SQL Azure Sync

Top Features Service scales as resources requirements grow (Windows Azure service) No-Code Sync Configuration (use Sync Framework if you wish to code). Schedule Sync Conflict Handling Logging and Monitoring Data sub-setting (@table in CTP2)

SQL Azure Reporting

SQL Azure Reporting Currently CTP – you can register Based on SQL Server Reporting Services 2008 R2 Exactly same report formats (no custom extensions yet) Use BI Development Studio No subscriptions Reports run on Windows Azure Platform Report Viewer control using remote processing mode V1 only reports against SQL Azure Database

SQL Azure Reporting vs SSRS Reporting Services SQL Azure Reporting Design Tool Business Intelligence Design Studio (BIDS) Report Builder Business Intelligence Design Studio (BIDS) SQL Server + others .e.g. Oracle, Teradata, PDW, etc. SQL Azure databases Data Sources Report Manager or SharePoint Display reports Render to multiple formats Subscriptions Scheduled delivery Embed reports into applications SQL Azure Development Portal Display reports Render to multiple formats Embed reports into applications Management & Delivery Developer Extensibility Custom data sources, assemblies, report items, authentication, etc. Extensibility not enabled. Security Model Supports Windows Authentication SQL Azure username/password

Demo

Futures

Scaling Database Applications Scale up Buy large-enough server for the job But big servers are expensive! Try to load it as much as you can But what if the load changes? Provisioning for peaks is expensive! Scale-out Partition data and load across many servers Small servers are cheap! Scale linearly Bring computational resources of many to bear Cluster of 100’s of little servers is very fast Load spikes not as problematic Load balancing across the entire cluster

SQL Azure Federations http://blogs.msdn.com/b/cbiyikoglu/ CREATE DATABASE SalesDB go CREATE FEDERATION CustFed (cid BIGINT RANGE) USE FEDERATION CustFed(cid=0) WITH FILTERING=OFF, RESET CREATE TABLE Customer( CustomerID bigint Primary Key, …) FEDERATED ON (cid = CustomerID) Go USE FEDERATION ROOT WITH RESET ALTER FEDERATION CustFed SPLIT AT (cid = 100) USE FEDERATION CustFed(cid=105) WITH FILTERING=ON, RESET Select * from customer Azure DB with Federation Root Federation Directory, Federation Users, Federation Scheme, … Federation “CustFed” (Federation Key: CustomerID) Member: PK [min, 100) AU PK=5 AU PK=25 AU PK=35 Member: PK [100, 488) AU PK=105 AU PK=235 AU PK=365 Connection Gateway Member: PK [488, max) AU PK=555 AU PK=2545 AU PK=3565 Sharded Application http://blogs.msdn.com/b/cbiyikoglu/ 25

Summary SQL Azure Database SQL Azure Data Sync SQL Azure Reporting hassle free fault tolerance with little to learn SQL Azure Data Sync Helping build geographically disperse or hybrid applications SQL Azure Reporting Reporting Services in the Cloud SQL Azure Federation Massive Scalability (Futures)

© 2008 Microsoft Corporation. All rights reserved © 2008 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista 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.