Moore’s Law on total CPU processing power holds but in parallel processing… CPU clock rate stalled… Because processors would melt… New CPU won’t run a short transaction much faster Meanwhile RAM.
Download ReportTranscript Moore’s Law on total CPU processing power holds but in parallel processing… CPU clock rate stalled… Because processors would melt… New CPU won’t run a short transaction much faster Meanwhile RAM.
Moore’s Law on total CPU processing power holds but in parallel processing… CPU clock rate stalled… Because processors would melt… New CPU won’t run a short transaction much faster Meanwhile RAM cost continues to drop 1000000 100000 10000 1000 100 10 1 1990 1991 1992 1993 1994 1994 1995 1996 1997 1998 1999 2000 2000 2001 2002 2004 2005 2007 2008 2009 2011 US$/GB $ per GB of PC Class Memory Principles Performance-critical data fits in memory Push decisions to compilation time Architectural Pillars Main-Memory Optimized Direct pointers to rows Indexes exist only in memory No buffer pool No write-ahead logging Stream-based storage T-SQL Compiled to Native Machine Code • T-SQL compiled to machine code leveraging VC compiler • Procedure and its queries, becomes a C function • Aggressive compiletime optimizations In-memory cache speed with capabilities of a database Queries & business logic run at nativecode speed • • • • Results • Conflicts are Rare • • • • Non-Blocking Execution Multi-version optimistic concurrency control, full ACID support Lock-free data structures No locks, latches or spinlocks, No I/O in transaction Transactions execute to completion without blocking Built-In SQL Server Integration • Same manageability, administration & development experience • Integrated queries & transactions • Integrated HA and backup/restore Hybrid engine and integrated experience SQL Server 2014 In-Memory OLTP aka Hekaton 7x faster performance Enables real-time updates Business challenge Online gaming company needed to support rapid growth by providing a faster, smoother gaming experience for more users, and speeding time to insights. Key benefits Scales to handle 250,000 player requests per second— up from 15,000 Reduced reporting time from 17 minutes to 3 seconds Before Gaming App 15,000 requests/sec With in-memory OLTP & ColumnStore Reporting Reports in 17 min Gaming App 250,000 requests/sec Reporting Reports in 3 sec 800% performance gains on the same HW 4x performance 40% cost 24x boost in throughput on the same configuration The Road Ahead: In Memory OLTP in SQL 2016 SQL Transaction log DelDel Tran1 (TS150) ) Tran1(TS150 Log in SQL Table Del Tran2 (TS 450) DelTran3 Tran3 Del (TS 250) (TS 250) Insert into Hekaton T1 Single Offline Checkpoint Thread disk SQL Transaction log DelDel Tran1 (TS150) ) Tran1(TS150 Offline Checkpoint Thread Log in SQL Table Del Tran2 (TS 450) DelTran3 Tran3 Del (TS250) 250) (TS Offline Checkpoint Thread disk Insert into Hekaton T1 Offline Checkpoint Thread Increasing Compatibility ALTER support ‐ ‐ ‐ ‐ ‐ Increasing Platform Abilities ‐ ‐ In-memory enhancements Operational analytics & enhanced performance SQL Server In-memory OLTP data warehouse 2-24 hrs ETL Fraud detected Real-time fraud detection 0100101010110 In-memory SQL Server ColumnStore OLTP In-Memory Technologies Overview for Microsoft SQL Server and Microsoft Azure Operational Analytics in SQL Server In-Memory OLTP – The Road Ahead ColumnStore Index: Microsoft SQL Server 2014 and Beyond In-Memory OLTP Monitoring & Troubleshooting Microsoft SQL Server 2014 In-Memory OLTP and Columnstore Customer Deployments Microsoft SQL Server Unplugged Other In-Memory OLTP Labs are available as well throughout the conference http://myignite.microsoft.com