Kevin Cox, SQLCAT kevincox@microsoft.com The SQL Server Customer Advisory Team (SQL CAT) represents the customer-facing resources from the SQL Server Product Group. SQL CAT is comprised of product and solution experts that regularly engage in the largest, most complex, and most unique customer deployments worldwide. http://www.microsoft.com/presspass/press/2010/sep10/09-01temenospr.mspx http://sqlcat.com //missioncritical – field resources The Customer Experience (CX) Design Wins Program is how the Microsoft Business Platform Division identifies and invests in strategic and large-scale customer projects with challenging, unique, and complex applications running on the Microsoft platform. optional optional optional/limited availability! optional Example (SQL Server Qualifying Criteria) 10+ TB DW, 3k tx/s OLTP, Large 500GB+ Cubes, Competitive migrations, Complex deployments, Server Consolidation (1000+) http://sqlserver/customers/customeradvisoryteam/default.aspx Category Largest single database Largest table Biggest total data 1 application Highest database transactions per second 1 db (from Perfmon) Fastest I/O subsystem in production (SQLIO 64k buffer) Fastest “real time” cube data load for 1TB Largest cube Metric 70 TB 20 TB 88 PB 130,000 18 GB/sec Millisecond latency 30 minutes 12 TB SQL Server Proof Points VLDB Mission Critical HA/DR Data Warehouse OLTP Large U.S. Financial Organization >1 peta-byte Critical operations with > 1,000,000,000 tx/day High-Volume Large U.S. Financial Organization 100% uptime (2008) 100% uptime (2009) 180M incremental fact rows/day > 125K tx/second < 50ms latency Centipede > peta-byte ≈ 1 trillion rows >80GB daily growth >50 TB by end of 2010 >90GB daily growth >10B rows in 1 table Mission Critical database > 4 TB Critical operations with > 15,000 users Critical operations at > 2200 facilities Mission Critical table > 1.9 billion rows 7x24x365 Synchronous Mirroring Solution >15B tx/yr SAP | geo-cluster > 5 miles No Data Loss | No Down Time Asynch Mirroring > 400 miles < 60 sec. recovery COMING SOON! (10+1 Clusters) 490TB 9 data nodes 70GB daily growth <= 3s query response (80%) 700 million rows/day Incremental growth 5TB Analysis Services data 50TB federated environment > 30,000 DB tx/Second > 3,000 tx/Second > 2B CLR calculations/day 12.4M users (2.7M regular) > 23k batch requests/sec 4.4M concurrent users 130M monthly users http://www.microsoft.com/casestudies/casestudy.aspx?casestud yid=49192 http://www.microsoft.com/casestudies/Case_Study_Detail.a spx?casestudyid=4000007013 http://www.baselinemag.com/article2/0,1540,2082921,00.as p • World’s biggest publicly listed online gaming platform • World’s biggest publicly listed online gaming platform • World’s leading provider of online Sports Betting • One of the largest Poker networks • Comprehensive range of Payment Service Providing • Integrated gaming portal - 22 languages, 25 core markets • More than 20 million registered customers • 1,500 employees • bwin builds on the strengths of the web in order to tie responsibility and gaming • 15 million page views and up to 980,000 users a day http://sqlcat.com/whitepapers/archive/2009/08/13/a-technical-case-study-fast-and-reliable-backupand-restore-of-a-vldb-over-the-network.aspx http://www.microsoft.com/casestudies/Case_Study_Detail.aspx?casestudyid=4000001470 DB size of the benchmark is about 800GB NEC Express 5800 64 core to host SQL Server. App servers are scale-out could be consolidated on few high end machines like DL980 or spread out on lower cost servers. Transactions/sec 5203 tps –business transactions. SQL Server transactions are much lower as they bundle multiple biz trans in single database tran. Clustering used for high availability. Log Shipping for Disaster Recovery. Repl is used in some deployments as way of reporting Asia Web Asia Web U.S. Web U.S. Web U.S. Web Web Publication Web Publication P2P Financial P2P Reference P2P Reference DW Asia Core 1 U.S. Core 1 P2P Financial P2P Financial P2P Reference P2P Financial P2P Financial P2P Reference P2P Financial P2P Financial Read Only P2P Reference P2P Reference P2P Financial P2P Reference Asia Core 2 U.S Core 2 P2P Reference Web Publication Web Publication Asia Web Asia Web Asia Web U.S. Web U.S. Web here www.sqlcat.com HA Technology Review RPO Solutions No Data Loss (RPO=0) Redundancy and Utilization Failover Hard-ware App Perf Impact Manageability Low Low Low * Low High Low * Low Low Low Cluster High*** Low *** Low*** Transactional Replication Low Low High Peer-Peer Replication Low Low High Log Shipping DBM Sync Async Failover Unit Inst DB Tab Auto Failover (RTO) Read Multiple * + ** Write Cost * Database Mirroring and Log Shipping can provide point in time read capability using STANDBY or database snapshots respectively ** Database Mirroring provides fastest failover to hot secondary *** Depends on SAN technology The key is to build a Balanced System without bottlenecks NIC 1 4 Network Server Memory 2 HBA 3 Disk Subsystem 5 SQL File Layout SQL Server is only part of the equation. Eco system needs to scale. Memory Front side bus contention increases w/ higher #CPUs LP 0 LP 1 LP 2 LP 3 LP 4 LP 5 Symmetric Multiprocessor Architecture LP 0 LP 2 LP 1 LP 3 Local Memory Access Memory Foreign Memory Access Non-Uniform Memory Access Foreign memory access > local memory access LP 6 LP 7 PDW Appliance Control Rack Compute Rack Compute Nodes Control Nodes Storage Nodes Database Servers Active / Passive SQL SQL Example Based on HP Architecture SQL Management Servers SQL SQL SQL SQL Backup Node SQL SQL SQL Spare Database Server SQL Dual Fiber Channel Dual Infiniband Landing Zone SQL Compute Nodes: HP DL380 G6 HP MSA Your input is important! https://www.MyTechReady.com For more information please refer to your Pocket Guide Speaker – Click Here to Launch Video