Vertical Scaling of Oracle 10g Performance on Red Hat Enterprise Linux 5 on Intel Xeon Based Servers. Version 1.0



Similar documents
Removing Performance Bottlenecks in Databases with Red Hat Enterprise Linux and Violin Memory Flash Storage Arrays. Red Hat Performance Engineering

SAS Business Analytics. Base SAS for SAS 9.2

Performance and Scalability of the Red Hat Enterprise Healthcare Platform

Deploying IBM Lotus Domino on Red Hat Enterprise Linux 5. Version 1.0

HP ProLiant BL460c takes #1 performance on Siebel CRM Release 8.0 Benchmark Industry Applications running Linux, Oracle

Scaling Microsoft Exchange in a Red Hat Enterprise Virtualization Environment

Performance brief for IBM WebSphere Application Server 7.0 with VMware ESX 4.0 on HP ProLiant DL380 G6 server

Oracle Database Scalability in VMware ESX VMware ESX 3.5

Cisco Prime Home 5.0 Minimum System Requirements (Standalone and High Availability)

HP ProLiant BL460c achieves #1 performance spot on Siebel CRM Release 8.0 Benchmark Industry Applications running Microsoft, Oracle

Performance and scalability of a large OLTP workload

Storage Management for the Oracle Database on Red Hat Enterprise Linux 6: Using ASM With or Without ASMLib

Marvell DragonFly. TPC-C OLTP Database Benchmark: 20x Higher-performance using Marvell DragonFly NVCACHE with SanDisk X110 SSD 256GB

Sizing guide for SAP and VMware ESX Server running on HP ProLiant x86-64 platforms

vrealize Business System Requirements Guide

TekSouth Fights US Air Force Data Center Sprawl with iomemory

Performance Characteristics of VMFS and RDM VMware ESX Server 3.0.1

HP reference configuration for entry-level SAS Grid Manager solutions

Red Hat Enterprise Linux: The ideal platform for running your Oracle database

HPSA Agent Characterization

FOR SERVERS 2.2: FEATURE matrix

N /150/151/160 RAID Controller. N MegaRAID CacheCade. Feature Overview

HP ProLiant BL685c takes #1 Windows performance on Siebel CRM Release 8.0 Benchmark Industry Applications

HP ProLiant DL380 G5 takes #1 2P performance spot on Siebel CRM Release 8.0 Benchmark Industry Applications running Windows

How System Settings Impact PCIe SSD Performance

Performance Comparison of Fujitsu PRIMERGY and PRIMEPOWER Servers

Sun 8Gb/s Fibre Channel HBA Performance Advantages for Oracle Database

Performance characterization report for Microsoft Hyper-V R2 on HP StorageWorks P4500 SAN storage

RED HAT ENTERPRISE VIRTUALIZATION PERFORMANCE: SPECVIRT BENCHMARK

Red Hat Network Satellite Management and automation of your Red Hat Enterprise Linux environment

Desktop virtualization: Implementing Virtual Desktop Infrastructure (VDI) with HP. What is HP Virtual Desktop Infrastructure?

Red Hat Satellite Management and automation of your Red Hat Enterprise Linux environment

High Performance SQL Server with Storage Center 6.4 All Flash Array

EMC Unified Storage for Microsoft SQL Server 2008

Virtual Compute Appliance Frequently Asked Questions

White Paper. SAP NetWeaver Landscape Virtualization Management on VCE Vblock System 300 Family

How To Store Data On An Ocora Nosql Database On A Flash Memory Device On A Microsoft Flash Memory 2 (Iomemory)

Performance Evaluation of VMXNET3 Virtual Network Device VMware vsphere 4 build

Red Hat Enterprise Linux: The Clear Leader for Enterprise Web Applications

Microsoft SharePoint Server 2010

RED HAT ENTERPRISE VIRTUALIZATION FOR SERVERS: COMPETITIVE FEATURES

EMC Backup and Recovery for Microsoft SQL Server

HP ProLiant DL585 G5 earns #1 virtualization performance record on VMmark Benchmark

Microsoft SharePoint Server 2010

HP ProLiant BL660c Gen9 and Microsoft SQL Server 2014 technical brief

Blueprints for Scalable IBM Spectrum Protect (TSM) Disk-based Backup Solutions

The 8Gb Fibre Channel Adapter of Choice in Oracle Environments

Red Hat Global File System for scale-out web services

Best Practices for Deploying SSDs in a Microsoft SQL Server 2008 OLTP Environment with Dell EqualLogic PS-Series Arrays

RAID 5 rebuild performance in ProLiant

7 Real Benefits of a Virtual Infrastructure

SUN ORACLE EXADATA STORAGE SERVER

Fusion iomemory iodrive PCIe Application Accelerator Performance Testing

KVM Virtualized I/O Performance

Maximum performance, minimal risk for data warehousing

8Gb Fibre Channel Adapter of Choice in Microsoft Hyper-V Environments

StarWind iscsi SAN: Configuring Global Deduplication May 2012

Hardware and Software Requirements. Release 7.5.x PowerSchool Student Information System

Minimize cost and risk for data warehousing

Windows Server 2008 R2 for Itanium-Based Systems offers the following high-end features and capabilities:

Optimizing SQL Server Storage Performance with the PowerEdge R720

SAP database backup and restore solutions for HP StorageWorks Enterprise Virtual Array using HP Data Protector 6.1 software

Oracle Database Deployments with EMC CLARiiON AX4 Storage Systems

QuickSpecs. What's New HP 750GB 1.5G SATA 7.2K 3.5" Hard Disk Drive. HP Serial-ATA (SATA) Hard Drive Option Kits. Overview

Enabling Technologies for Distributed and Cloud Computing

RED HAT ENTERPRISE LINUX 7

ARCHITECTING COST-EFFECTIVE, SCALABLE ORACLE DATA WAREHOUSES

Red Hat enterprise virtualization 3.0 feature comparison

HP and Mimosa Systems A system for archiving, recovery, and storage optimization white paper

EMC Backup and Recovery for Microsoft SQL Server

Cisco Unified Computing System and EMC VNX5300 Unified Storage Platform

RED HAT ENTERPRISE VIRTUALIZATION FOR SERVERS: PRICING & LICENSING GUIDE

An Oracle White Paper April 2012; Revised July Improving Data Center Infrastructure with Oracle s Sun x86 Systems

Using VMware VMotion with Oracle Database and EMC CLARiiON Storage Systems

System and Storage Virtualization For ios (AS/400) Environment

Technical Paper. Moving SAS Applications from a Physical to a Virtual VMware Environment

How To Compare Two Servers For A Test On A Poweredge R710 And Poweredge G5P (Poweredge) (Power Edge) (Dell) Poweredge Poweredge And Powerpowerpoweredge (Powerpower) G5I (

Vormetric and SanDisk : Encryption-at-Rest for Active Data Sets

PARALLELS SERVER 4 BARE METAL README

Re-Hosting Mainframe Applications on Intel Xeon Processor-based Servers

Oracle TimesTen In-Memory Database on Oracle Exalogic Elastic Cloud

QuickSpecs. HP SATA Hard Drives. Overview

QuickSpecs. HP Serial-ATA (SATA) Entry (ETY) and Midline (MDL) Hard Drive Option Kits. Overview

An Oracle White Paper August Oracle WebCenter Content 11gR1 Performance Testing Results

<Insert Picture Here> Introducing Oracle VM: Oracle s Virtualization Product Strategy

3 Red Hat Enterprise Linux 6 Consolidation

An Oracle White Paper August Oracle VM 3: Server Pool Deployment Planning Considerations for Scalability and Availability

ACANO SOLUTION VIRTUALIZED DEPLOYMENTS. White Paper. Simon Evans, Acano Chief Scientist

Delivers high performance, reliability, and security. Is certified by the leading hardware and software vendors

Virtualization Performance on SGI UV 2000 using Red Hat Enterprise Linux 6.3 KVM

Virtuoso and Database Scalability

HP ProLiant Essentials Performance Management Pack 3.0 Support Matrix

Save up to 85% on Your Oracle Costs

HP StorageWorks Entry-level Enterprise Backup Solution

HP PCIe IO Accelerator For Proliant Rackmount Servers And BladeSystems

Red Hat Enterprise Linux and management bundle for HP BladeSystem TM

Handling Multimedia Under Desktop Virtualization for Knowledge Workers

HP Intelligent Management Center User Access Management Software

Server Migration from UNIX/RISC to Red Hat Enterprise Linux on Intel Xeon Processors:

Transcription:

Vertical Scaling of Oracle 10g Performance on Red Hat Enterprise Linux 5 on Intel Xeon Based Servers Version 1.0 March 2009

Vertical Scaling of Oracle 10g Performance on Red Hat Enterprise Linux 5 on Inel Xeon Based Servers 1801 Varsity Drive Raleigh NC 27606-2072 USA Phone: +1 919 754 3700 Phone: 888 733 4281 Fax: +1 919 754 3701 PO Box 13588 Research Triangle Park NC 27709 USA Linux is a registered trademark of Linus Torvalds. Red Hat, Red Hat Enterprise Linux and the Red Hat "Shadowman" logo are registered trademarks of Red Hat, Inc. in the United States and other countries. All other trademarks referenced herein are the property of their respective owners. 2009 by Red Hat, Inc. This material may be distributed only subject to the terms and conditions set forth in the Open Publication License, V1.0 or later (the latest version is presently available at http://www.opencontent.org/openpub/). The information contained herein is subject to change without notice. Red Hat, Inc. shall not be liable for technical or editorial errors or omissions contained herein. Distribution of modified versions of this document is prohibited without the explicit permission of Red Hat Inc. Distribution of this work or derivative of this work in any standard (paper) book for commercial purposes is prohibited unless prior permission is obtained from Red Hat Inc. The GPG fingerprint of the security@redhat.com key is: CA 20 86 86 2B D6 9D FC 65 F6 EC C4 21 91 80 CD DB 42 A6 0E www.redhat.com 2

Table of Contents 1. Introduction & Executive Summary... 4 2. Performance Testing Methodology... 5 3. Oracle OLTP Performance & Scaling on RHEL 5 / Harpertown (Intel E5440)... 6 3.1 Configuration... 6 3.2 Results... 6 4. Oracle OLTP Performance & Scaling on RHEL 5 / Tigerton (Intel X7350)... 7 4.1 Configuration... 7 4.2 Results... 7 5. Oracle OLTP Performance & Scaling on RHEL 5 / Dunnington (Intel X7600)... 8 5.1 Configuration... 8 5.2 Results... 8 6. Conclusions... 9 7. References... 9 3 www.redhat.com

1. Introduction & Executive Summary Customers seriously considering migrating mission-critical applications from UNIX to Red Hat Enterprise Linux (RHEL) are demanding up to date vertical scaling data for Oracle performance. This is an update of an earlier study which demonstrated scaling up to 8 cores. The goal of this study was to characterize the performance and vertical scaling of Oracle OLTP performance on RHEL 5 running on Intel Xeon based servers (up to 24 cores). The OLTP workload was used because it represents a very common type of database workload which exercises both the memory and I/O sub-systems. The performance metric is TPM (Transactions Per Minute). Since this TPM performance metric is for a synthetic benchmark workload, the actual TPM for a real application will depend on the characteristics of the application. Figure 1 illustrates the excellent scaling achieved up to 24 cores. Figure 1 www.redhat.com 4

2. Performance Testing Methodology Database performance and scalability often depend on many factors. However, it is impractical to try and collect data for all values of all the factors that can influence Oracle performance on RHEL. In order to get some useful information out to our sales, support and consultants in the field in a timely manner some judgment must be exercised in selecting the subset of the data that can be collected with the time and resources available. Specifically selected were: A. Workload Types: Online Transaction Processing (OLTP) Decision Support System (DSS) The OLTP workload was used because it represents a very common type of database workload which exercises both the memory and I/O sub-systems. DSS workloads are less common and require large storage sub-systems. Since access to such storage subsystems was limited, we limited this version of the study to OLTP workloads. B. File Systems: EXT3 GFS NFS RAW device pseudo-files The EXT3 file system was selected as the basis for most of the experiments. EXT3 is a journalled file system and is the default system on Red Hat Enterprise Linux. Using a file system and/or LVM volumes for the Oracle databases maps to a large percent of users of Oracle on RHEL. C. I/O Modes: Synchronous I/O Asynchronous I/O (AIO) Direct I/O (DIO) AIO with DIO (AIO + DIO) Raw I/O In previous experiments it was observed that AIO+DIO consistently performed better than Synch I/O, Asynch I/O (AIO) and Direct I/O (DIO) for OLTP workload. So, studying AIO+DIO would yield the upper bound OLTP performance and scaling. AIO+DIO is what we recommend to our customers for OLTP workloads. D. X86_64 Servers: Intel Xeon AMD Opteron Processors from Intel s Harpertown, Tigerton and Dunnington families were used. The enterprise customers, simulated in these experiments, used the above OLTP workload with an average database size between 50-100 GB. 5 www.redhat.com

3. Oracle OLTP Performance & Scaling on RHEL 5 / Harpertown (Intel E5440) 3.1 Configuration Dual Socket, Quad-Core (Total of 8 Cores) Processor Intel Xeon E5440 (Harpertown) 2.83 GHz 8 GB RAM EqualLogic PS3800XV iscsi Array Storage 4.8 TB (16 x 300 GB, 15 K RPM SAS Drives) Operating System Red Hat Enterprise Linux (RHEL) 5.1 File System Database EXT3 Oracle 10g R2 3.2 Results Figure 2 www.redhat.com 6

4. Oracle OLTP Performance & Scaling on RHEL 5 / Tigerton (Intel X7350) 4.1 Configuration Quad Socket, Quad-Core (Total of 16 Cores) Processor Intel Xeon X7350 (Tigerton) 2.93 GHz 32 GB RAM HP StorageWorks 4400 Enterprise Virtual Array (EVA) Storage 5.6 TB (20 x 280 GB Drives) Operating System Red Hat Enterprise Linux (RHEL) 5.3 File System Database EXT3 Oracle 10g R2 4.2 Results Figure 3 7 www.redhat.com

5. Oracle OLTP Performance & Scaling on RHEL 5 / Dunnington (Intel X7600) 5.1 Configuration Quad Socket, Hex-Core (Total of 24 Cores) Processor Intel Xeon X7600 (Dunnington) 2.66 GHz 32 GB RAM HP StorageWorks 2000 Modular Smart Array (MSA) Storage 1.716 TB (12 x 146 GB Drives) for Data + 2 x 80 GB Fusion IO SSD for Logs Operating System Red Hat Enterprise Linux (RHEL) 5.3 File System Database EXT3 Oracle 10g R2 5.2 Results Figure 4 www.redhat.com 8

6. Conclusions Database OLTP workloads represent a large portion of enterprise database customers real environments today. Chip vendors are driving scalability inovation through lager processor counts and cores/chip. As this trend continues to increase from dual-to-quad, quad-to-hex core on x86_64 servers, Open source Red Hat Enterprise Linux will continue to optimize the OS for database scalability for enterprise applications like Oracle such that users derive incremental improvement in real customer environments as cpu-cores increase. While the hardware design ultimately determines the position of the scale curve, the OS and it capabilities effect the slope of the curve. This study demonstrates that given adequate I/O bandwidth, Oracle database OLTP performance scales very well up to 24 cores when Red Hat Enterprise Linux (RHEL) 5 on Intel Xeon servers. In the future, we expect to demonstrate vertical scaling of database OLTP performance to servers with: even larger number of cores / hyper-threads newer chip technology, e.g., Intel Nahalem 7. References 1. Red Hat Reference Architecture: Tuning & Optimizing Red Hat Enterprise Linux (RHEL) for Oracle 9i and 10g Databases. Version 1.1, November 2007. https://bu-platform.devel.redhat.com/bu/prgrms/refarch/oracle-tuning-guide%201-1%2011-28-2007.pdf 2. Red Hat Reference Architecture: Performance & Scalability of Virtualized Oracle 10g Servers on Red Hat Enterprise Linux, Version 1.0, June 2008. https://bu-platform.devel.redhat.com/bu/prgrms/refarch/oracle-10g-recommendationsv1.2.pdf 3. Red Hat Reference Architecture: Oracle 10g Server on Red Hat Enterprise Linux 5 Deployment Recommendations, Version 1.2, November 2008. https://buplatform.devel.redhat.com/bu/prgrms/refarch/virtualized%20oracle%20server%20perf ormance%20v1.pdf 9 www.redhat.com