Time Breakdowns A novel way to analyze performance Problems

Size: px
Start display at page:

Download "Time Breakdowns A novel way to analyze performance Problems"

Transcription

1 Time Breakdowns A novel way to analyze performance Problems Roy Cecil IBM Session Code: 3 16 April 2014 Platform: LUW

2 2 Agenda Introduction Amdahl's law Time Breakdowns Examples Bottlenecks & Little law

3 Every performance problem is an optimization problem! #IDUG

4 What are we optimizing? Business Metric Transactions processed per second Number of orders processed per day Number of containers moved per hour at a dock Defects in a factory line In the context of DB2 Transactions per second (Throughput) Number of Reports processed per second in a Data Warehouse Response time of shopping cart transactions

5 What are the attributes of a business metric? Is it quantifiable? Under identical conditions is it repeatable? Is the problem within the accuracy with which the metric is expressed? If being compared across systems are they comparable?

6 Setting Goals Do all stakeholders sign off? #IDUG

7 7 Agenda Introduction Amdahl's law Time Breakdowns Examples Bottlenecks & Little law

8 Amdahl's law : The speedup achievable from an improvement to a computation that affects a proportion P of that computation where the improvement has a speedup of Sis given by 1 / [ ( 1 P ) + P/S ] Q P = 0.8 Q P/2 =0.4 Q = 1- P 1.67 x ie 67% overall improvement Q/2 P = x ie only 11% improvement It doesnt pay a lot to improve breakdowns that consume a negligible fraction of the overall elapsed time.

9 Law of Diminishing return : If the response time is broken down to various sub units and one picked optimally what to improve then one would see monotonically decreasing improvement as one improves. Time breakdown of a Transaction Optimal A B -> A -> C -> D B C D Sub-optimal D -> A -> C -> B

10 10 Agenda Introduction Amdahl's law Time Breakdowns Examples Bottlenecks & Little law

11 Time Breakdowns The fundamental unit of work in DB2 is the Transactions. Most often the Business metric being optimized is directly translated to Database throughput. Throughput is defined as the number of transactions that completedinagivenunitoftime. An alternate view of the Business metric that derives directly from the Throughput is the Average Response Time of the Transactions. Throughput ( T ) = 1 / R [ R is the Average Response Time of the transactions]

12 Time Breakdowns To obtain the Time breakdowns, the response time of the individual transactions or the total time spent in DB2 are broken down into the time spent in various components while processing that transaction. Coupled with Amdahl's law we should have a better understanding of what component to optimize to be able to get a greater throughput

13 Two dimensions of viewing Time breakdowns System wide Dimension Used to see system wide time breakdowns. Can also be used to see the time breakdowns per workload. Useful if there is a systemic performance problem rather than a problem with specific activities. Specific Activity Dimension Can be used to see time breakdowns of specific activities. For example time breakdowns of a Unit of Work Rolled up into System wide time breakdowns.

14 Three aspects of time monitored in DB2 Wait Times Waiting for latches or Locks Waiting for incoming client requests Reading or Writing from Bufferpool Component processing Times Committing or rolling back transactions Compiling SQL Component elapsed times Total Commit time Total compile time

15 Some facts to remember Time Breakdown monitor elements The Wait time elements and Component processing times are complementary to each other. Σ ( wait times ) + Σ ( component processing time ) total_rqst_time The small difference arises from components not monitored in DB2. This is sufficient to provide good insight into problem areas. Component elapsed time includes component processing times. Component elapsed times already includes wait times Not meaningful to use component times and wait times as that would amount to double counting.

16 Breaking down Total Elapsed Time System Dimension Total Elapsed Time ( wall clock ) is the sum of CLIENT_IDLE_WAIT_TIME and TOTAL_RQST_TIME CLIENT_IDLE_WAIT_TIME and TOTAL_RQST_TIME should be normalized by the number of connections. CLIENT_IDLE_WAIT_TIME is the amount of time DB2 spent waiting for work from Clients. TOTAL_RQST_TIME is the amount of time spent doing useful work in DB2.

17 Breaking down Total Elapsed Time - Example A snippet from MON_GET_WORKLOAD. Monitoring Interval = 200 seconds ; Number of clients = 40 Both TOTAL_RQST_TIME and CLIENT_IDLE_WAIT_TIME is expressed in milliseconds The elapsed time is the same as the Monitoring interval. TOTAL_RQST_TIME CLIENT_IDLE_WAIT_TIME TOTAL_RQST_TIME = CLIENT_IDLE_WAIT_TIME = Elapsed Time = ( ) / ( 40 * 1000 ) = 200 TOTAL_RQST_TIME = / ( 40 * 1000) % TOTAL_RQST_TIME = / 200 * 100 = 43.48%

18 Breaking down Total Request Time Time spent in DB2 Total Time spent processing Requests can be broken down into Total Wait time events in DB2 Total Compile Times Explicit ( initiated by user prep/bind/rebind etc ) Implicit ( not initiated by user ) Total user code in routines Total Section processing times Total Commit & Rollback times Total Runstat, Reorg and Load processing times Total Connection Request times Others

19 Performance Objective Reduce Wait times #IDUG

20 Significant Wait times Performance opportunities Lock Wait Times. Log Buffer and Log Disk wait times. Network Wait Times I/O Wait times TCP / IPC / FCM Send & Receive wait times Pool Read/Write times Direct Read/Write times Prefetch wait times. CF ( Caching Facility in DB2 purescale ) wait times. Reclaim Wait Times Latch Wait times Others.

21 Understanding Lock Wait times. The average time the applications spend waiting for locks is given by lock_wait_time / lock_waits Are the waits due to Lock Escalations ( lock_escals )? Increase LOCKLIST / MAXLOCKS mon_get_locks and mon_get_appl_lockwait can be used to diagnose Lock wait problems and help break down the picture to individual packages/statements. General good ideas for reducing lock wait Use the lowest isolation level that meets business requirement Commit optimally. Strive for highly selective access plans ( indexed vs scanned ) If we see a lot of row locks with reads ( S locks ) waiting on writes ( X locks ) enabling CUR_COMMIT database configuration can help when CS isolation is in force.

22 Log Buffer Wait times. A large wait by the agent for Log buffer is indicative of a higher rate at which the buffer fills up before being flushed to the disk. num_log_buffer_full need to be checked to see if applications cannot insert into the log buffer because the full buffer is being flushed to the disk. If num_log_buffer_full is zero a very probable cause for this wait is implicit commits. Tuning the LOGBUFSZ will ensure that there is sufficient space in the log buffer between successive log flushes. Ideally the value for log_buffer_wait_time should be zero. Increasing the LOGBUFSZ when log_buffer_wait_time or num_log_buffer_full have a value of zero could have adverse impact on the performance of the system.

23 Log Disk Wait times. A higher log disk wait time can limit the transaction rateand often would require a more faster dedicated I/O subsystem Log disks writes are sequential and would benefit from Faster disks available in the market with higher throughput. SSD's have higher throughput than spinning disks. Striping the log device across multiple disks will improve the write response times. Use dedicated filesystems separate from Tablespaces and other applications. Enabling write cache on the storage controller will improve log write times. Not inlining lobs that are updated would help reduce impact on the Log file system. Instead these can be cached in the Filesystem. Evaluate possibility of using 'Not logged initially' for bulk operations. Care should be taken to have a minimum amount of rollbacks. Rollbacks would require random disk access and could degrade the performance of the system.

24 Network Wait times. FCM send and receive wait times are incurred when a query is run in a DPF environment or when intra query parallelism is turned on. A small amount of FCM wait times will also be observed in purescale owing to member to member chatter Tuning FCM buffers ( fcm_num_buffers ) and FCM channels ( fcm_num_channels ) can help alleviate the problem if there are insufficient FCM buffers or channels configured. TCP/IP and IPC send and receive wait times are incurred when large amount of data needed to be send to the client. Eg. When entire tables are fetched from the database to be cached in the application layers. In the absence of such use non-selective data fetch use cases consider tuning the network stack or improving the network bandwidth between the agent and the DB2 server.

25 I/O Bottlenecks Pool Read time A higher value of pool_read_time is indicative of agents waiting for data to be fetched from the physical storage device into the bufferpool Can be broken down with MON_GET_ACTIVITY/MON_GET_PKG_CACHE_STMT into specific statements. Are you getting an unwanted tablescan? Monitor rows_read >> num_executions? Confirm plan with db2expln or db2exfmt Query is repeated, not ad hoc Are statistics out of date? Old/inaccurate statistics can trigger a tablescan Is the table sufficiently indexed? The DB2 Design Advisor may be able to help Other possibilities Is Materialized Query Table or Multi Dimensional clustering suitable?

26 I/O Bottlenecks Pool write time A higher value of pool_write_time is indicative of a high level of page cleaning activity. Consider tuning SOFTMAX to a higher value if your Recovery Time Objectives ( RTO ) is not super aggressive. Note: From version 10.5 page_age_target_mcr or page_age_target_gcr replaces SOFTMAX parameter. Analyze the iostat output to see if the disk service times are higher. And consider adding more spindles or faster disks. Tuning the NUM_IO_CLEANERS can help if the disk service times are not high.

27 I/O Bottlenecks Direct Read time and Direct write time MON_GET_ACTIVITY/mon_get_package_cache_stmt table function can be used to break this down to specific statements. LOB inlining can be used to improve the performance of data access by caching frequently accessed LOBS in bufferpools. Moving tables with LOBS to its own table-spaces and enabling filesystem caching can also improve the performance of systems bottle-necked on Direct read/write operations.

28 I/O Bottlenecks Prefetch wait times. Consider tuning the NUM_IO_SERVERS by potentially adding more prefetchers who can prefetch data into the bufferpool. Consider the possibility of tuning the PREFETCHSIZE of the tablespaces. Data is prefetched in size of PREFETCHSIZE pages. Smaller bufferpools can cause data to be replaced from the bufferpool requiring more cases when pages accessed are not in the bufferpool. This can cause prefetcher to work harder. Keep up with improvements in Prefetching in DB2! DB2 LUW v10.5 has specific improvements in the prefetching area. Smart prefetching uses either readahead or sequential prefetching depending on the clustering of data/indexes. Reduces the need to do expensive operations like reorgs.

29 Latches What are they? Latches are internal DB2 locks used to serialize access to shared data structures. Eg. Pages. MON_GET_EXTENDED_LATCH_WAIT table function can be used to get an understanding of the latch waits. A higher latch wait could be an indirect manifestation of a poorly designed application. For eg. A Page latch is obtained before making a change to the data page. If there are multiple EDU's making concurrent changes to a hot page the access is serialized and could result in latch contention. If the contented page is an index page, using RANDOM indexes and partitioned tables might help.

30 CF Wait times & Reclaim Times Check if the IB/10GE cards are saturated Consider adding more I/O channels on the CF machines if they are saturated IF we see high reclaim times, consider using Member sub-setting / partitioned workloads with EHL ( Explicit Hierarchical Locking ) CURRENT_MEMBER registry variables can be used to improve contention issues. Note : CF stands for Caching Facility in DB2 purescale

31 Activity Dimension There are two cases under which we want to look at activity dimension When there is a specific wait time in the system dimension that we need to break down into specific activities to understand the cause of the wait time. ( Eg. Breaking down the pool_read_time to understand hot tables and statements ) When we are interested in improving the performance of specific activities. ( Eg. Improve the performance of Order entry transactions in an order processing system ) The principles of breaking down the activity to specific sub activities is very similar to what we saw in the system dimension.

32 Monitoring Table functions that give Time spent breakdowns System Dimension MON_GET_SERVICE_SUBCLASS MON_GET_WORKLOAD Activity Dimension MON_GET_CONNECTION MON_GET_UNIT_OF_WORK MON_GET_PACKAGE_CACHE_STMT

33 33 Agenda Introduction Amdahl's law Time Breakdowns Examples Bottlenecks & Little law

34 Case Study We have a workload that is performing poorly and we want to use the above principles to understand and try and improve the performance. The workload we chose is TPCC benchmark and we used data from the MON_GET_WORKLOAD to understand the time breakdowns from the system dimension.

35 Example of a time-spent breakdown TOTAL_APP_RQST_TIME Throughput = Transactions per Second TOTAL_WAIT_TIME Throughput = Transactions per Second 30% 39% TOTAL_WAIT_TIME OTHERS POOL_READ_TIME POOL_WRITE_TIME OTHERS 61% 0% 70% Workload is TPCC

36 Example of a time-spent breakdown TOTAL_APP_RQST_TIME T hroughput = T ransact ions per Second TOTAL_WAIT_TIME T hroughput = T ransact ions per Second 43% TOTAL_WAIT_TIME OTHERS 32% POOL_READ_TIME POOL_WRITE_TIME OTHERS 57% 68% 0% Culprit! Number of Bufferpool Pages Old Run = New Run =

37 Percentages dont tell the whole story! TOTAL_WAIT_TIME T hroughput = T ransact ions per Second TOTAL_WAIT_TIME T hroughput = T ransact ions per Second % POOL_READ_TIME POOL_WRITE_TIME OTHERS % POOL_READ_TIME POOL_WRITE_TIME OTHERS % % % % Applying Amdahl's Law Improvement = 1/ [ ( 1-P) + P/S ] P ( %age TOTAL_WAIT_TIME) = 0.61 S ( Speedup ) = / = 2.13 Improvement = 47% which matches the TPS improvement we got.

38 38 Agenda Introduction Amdahl's law Time Breakdowns Examples Bottlenecks & Little law

39 39 Bottlenecks What are they?

40 40 Little's Law Littles's law : At steady state, the average number of outstanding requests to a system is equal to the effective arrival rate multiplied by the average service time. Note : At steady state, the arrival rate is equal to the throughput of the system.

41 41 Where is the bottleneck? Webserver L = λ * S L1 = λ1 1 * S1 L2 = λ2 2 * S2 DB2 Total Queue Length ( L ) = Queue Length at Webserver ( L1 ) +Queue Length at DB2 ( L2 ) Bottleneck is the component with the slowest service time.

42 How to compute Little's Law from DB2 Monitoring Data 42 Average Service Time ( S ) = TOTAL_RQST_TIME ( System Dimension ) / ( TOTAL_APP_COMMITS + ROLLBACK_SQL_STMTS) ( in milliseconds ) Arrival Rate ( λ )= ( TOTAL_APP_COMMITS + ROLLBACK_SQL_STMTS) / Monitoring Interval From Little's Law we have L = λ * S Queue Length = TOTAL_RQST_TIME / Monitoring interval This is the same as the number of applications in 'UOW Executing' state from db2pd -applications at any given time.

43 How to compute Little's Law from DB2 Monitoring Data 43 If the Queue Length at DB2 is far less than the Queue lengths elsewhere in the system, DB2 is not the bottleneck. DB2 itself can be decomposed into multiple queuing systems where at each category of wait_time we can imagine a queue. Extending the above formula for queue length we can derive the queue length of specific wait events ( I/O, network, prefetch etc ) by taking the ratio of specific wait times to the monitoring interval. Queue Length ( L ) = TOTAL_RQST_TIME / Monitoring interval Monitoring interval = Σ ( agent_wait_time + lock_wait_time + ) / From the above formula is easy to see that the length of the queue is proportional to the wait times in specific components.

44 44 Questions?

45 Roy Cecil IBM Session 3 Time Breakdowns A novel way to analyze performance Problems Please fill out your session evaluation before leaving!

IBM DB2 LUW Performance Tuning and Monitoring for Single and Multiple Partition DBs

IBM DB2 LUW Performance Tuning and Monitoring for Single and Multiple Partition DBs IBM DB2 LUW Performance Tuning and Monitoring for Single and Multiple Partition DBs Day(s): 5 Course Code: CL442G Overview Learn how to tune for optimum the IBM DB2 9 for Linux, UNIX, and Windows relational

More information

DB2 for LUW Performance: You ve got (frequently asked) questions, we ve got answers!

DB2 for LUW Performance: You ve got (frequently asked) questions, we ve got answers! DB2 for LUW Performance: You ve got (frequently asked) questions, we ve got answers! Steve Rees IBM Canada Ltd. Session Code: D11 Wednesday November 10, 13:30 Platform: Linux / Unix / Windows Abstract

More information

DB2 Performance Essentials

DB2 Performance Essentials DB2 Performance Essentials Philip K. Gunning Certified Advanced DB2 Expert Consultant, Lecturer, Author DISCLAIMER This material references numerous hardware and software products by their trade names.

More information

Common DB2 Customer Issues. As seen by DB2-LUW support

Common DB2 Customer Issues. As seen by DB2-LUW support Common DB2 Customer Issues As seen by DB2-LUW support Hans Siebrand (hans_siebrand@kr.ibm.com) Jin Hwan Hyun (jhhyun2@kr.ibm.com) Information Management July 12, 2012 DB2 Support 3 levels 1 First contact

More information

Best Practices. Deploying Optim Performance Manager in large scale environments. IBM Optim Performance Manager Extended Edition V4.1.0.

Best Practices. Deploying Optim Performance Manager in large scale environments. IBM Optim Performance Manager Extended Edition V4.1.0. IBM Optim Performance Manager Extended Edition V4.1.0.1 Best Practices Deploying Optim Performance Manager in large scale environments Ute Baumbach (bmb@de.ibm.com) Optim Performance Manager Development

More information

Tuning DBM and DB Configuration Parameters

Tuning DBM and DB Configuration Parameters DB2 for Linux, UNIX, Windows Tuning DBM and DB Configuration Parameters Philip K. Gunning DB2 Master Practitioner Gunning Technology Solutions, LLC 21 September 2005 DB2 is a registered trademark of IBM

More information

What Developers must know about DB2 for z/os indexes

What Developers must know about DB2 for z/os indexes CRISTIAN MOLARO CRISTIAN@MOLARO.BE What Developers must know about DB2 for z/os indexes Mardi 22 novembre 2016 Tour Europlaza, Paris-La Défense What Developers must know about DB2 for z/os indexes Introduction

More information

Developing SQL Databases (762)

Developing SQL Databases (762) Developing SQL Databases (762) Design and implement database objects Design and implement a relational database schema Design tables and schemas based on business requirements, improve the design of tables

More information

Performance 101 for DB2 for LUW

Performance 101 for DB2 for LUW Performance 101 for DB2 for LUW A PDF of these slides can be downloaded from: ibm.com/developerworks/data/events/idmbriefings.html Jeff M. Sullivan DB2 on LUW and DB2 on z/os I.T. Specialist Optim Technical

More information

Performance Tuning for MDM Hub for IBM DB2

Performance Tuning for MDM Hub for IBM DB2 Performance Tuning for MDM Hub for IBM DB2 2012 Informatica Corporation. No part of this document may be reproduced or transmitted in any form, by any means (electronic, photocopying, recording or otherwise)

More information

Heckaton. SQL Server's Memory Optimized OLTP Engine

Heckaton. SQL Server's Memory Optimized OLTP Engine Heckaton SQL Server's Memory Optimized OLTP Engine Agenda Introduction to Hekaton Design Consideration High Level Architecture Storage and Indexing Query Processing Transaction Management Transaction Durability

More information

Exadata X3 in action: Measuring Smart Scan efficiency with AWR. Franck Pachot Senior Consultant

Exadata X3 in action: Measuring Smart Scan efficiency with AWR. Franck Pachot Senior Consultant Exadata X3 in action: Measuring Smart Scan efficiency with AWR Franck Pachot Senior Consultant 16 March 2013 1 Exadata X3 in action: Measuring Smart Scan efficiency with AWR Exadata comes with new statistics

More information

Performance Monitoring

Performance Monitoring Performance Monitoring Performance Monitoring Goals Monitoring should check that the performanceinfluencing database parameters are correctly set and if they are not, it should point to where the problems

More information

Application Development Best Practice for Q Replication Performance

Application Development Best Practice for Q Replication Performance Ya Liu, liuya@cn.ibm.com InfoSphere Data Replication Technical Enablement, CDL, IBM Application Development Best Practice for Q Replication Performance Information Management Agenda Q Replication product

More information

Outline. Database Tuning. Ideal Transaction. Concurrency Tuning Goals. Concurrency Tuning. Nikolaus Augsten. Lock Tuning. Unit 8 WS 2013/2014

Outline. Database Tuning. Ideal Transaction. Concurrency Tuning Goals. Concurrency Tuning. Nikolaus Augsten. Lock Tuning. Unit 8 WS 2013/2014 Outline Database Tuning Nikolaus Augsten University of Salzburg Department of Computer Science Database Group 1 Unit 8 WS 2013/2014 Adapted from Database Tuning by Dennis Shasha and Philippe Bonnet. Nikolaus

More information

MySQL Performance Optimization and Troubleshooting with PMM. Peter Zaitsev, CEO, Percona

MySQL Performance Optimization and Troubleshooting with PMM. Peter Zaitsev, CEO, Percona MySQL Performance Optimization and Troubleshooting with PMM Peter Zaitsev, CEO, Percona In the Presentation Practical approach to deal with some of the common MySQL Issues 2 Assumptions You re looking

More information

CA Unified Infrastructure Management Snap

CA Unified Infrastructure Management Snap CA Unified Infrastructure Management Snap Configuration Guide for DB2 Database Monitoring db2 v4.0 series Copyright Notice This online help system (the "System") is for your informational purposes only

More information

Database Management and Tuning

Database Management and Tuning Database Management and Tuning Concurrency Tuning Johann Gamper Free University of Bozen-Bolzano Faculty of Computer Science IDSE Unit 8 May 10, 2012 Acknowledgements: The slides are provided by Nikolaus

More information

WHITE PAPER. VERITAS Database Edition 1.0 for DB2 PERFORMANCE BRIEF OLTP COMPARISON AIX 5L. September

WHITE PAPER. VERITAS Database Edition 1.0 for DB2 PERFORMANCE BRIEF OLTP COMPARISON AIX 5L. September WHITE PAPER VERITAS Database Edition 1.0 for DB2 PERFORMANCE BRIEF OLTP COMPARISON AIX 5L September 2002 1 TABLE OF CONTENTS Introduction...3 Test Configuration...3 Results and Analysis...4 Conclusions...6

More information

IT Best Practices Audit TCS offers a wide range of IT Best Practices Audit content covering 15 subjects and over 2200 topics, including:

IT Best Practices Audit TCS offers a wide range of IT Best Practices Audit content covering 15 subjects and over 2200 topics, including: IT Best Practices Audit TCS offers a wide range of IT Best Practices Audit content covering 15 subjects and over 2200 topics, including: 1. IT Cost Containment 84 topics 2. Cloud Computing Readiness 225

More information

Segregating Data Within Databases for Performance Prepared by Bill Hulsizer

Segregating Data Within Databases for Performance Prepared by Bill Hulsizer Segregating Data Within Databases for Performance Prepared by Bill Hulsizer When designing databases, segregating data within tables is usually important and sometimes very important. The higher the volume

More information

MySQL Performance Optimization and Troubleshooting with PMM. Peter Zaitsev, CEO, Percona Percona Technical Webinars 9 May 2018

MySQL Performance Optimization and Troubleshooting with PMM. Peter Zaitsev, CEO, Percona Percona Technical Webinars 9 May 2018 MySQL Performance Optimization and Troubleshooting with PMM Peter Zaitsev, CEO, Percona Percona Technical Webinars 9 May 2018 Few words about Percona Monitoring and Management (PMM) 100% Free, Open Source

More information

An A-Z of System Performance for DB2 for z/os

An A-Z of System Performance for DB2 for z/os Phil Grainger, Lead Product Manager BMC Software March, 2016 An A-Z of System Performance for DB2 for z/os The Challenge Simplistically, DB2 will be doing one (and only one) of the following at any one

More information

Optimising Insert Performance. John Campbell Distinguished Engineer IBM DB2 for z/os Development

Optimising Insert Performance. John Campbell Distinguished Engineer IBM DB2 for z/os Development DB2 for z/os Optimising Insert Performance John Campbell Distinguished Engineer IBM DB2 for z/os Development Objectives Understand typical performance bottlenecks How to design and optimise for high performance

More information

Accelerate Your DB2 Business On Demand Autonomically!

Accelerate Your DB2 Business On Demand Autonomically! Session: C5 Accelerate Your DB2 Business On Demand Autonomically! Scott Hayes DBI (www.database-brothers.com) Nov 06, 2007 11:45 a.m. 12:45 p.m. (C5) Platform: DB2 for Linux, UNIX, Windows 1 Agenda A bit

More information

IBM Tivoli Storage Manager for Windows Version Installation Guide IBM

IBM Tivoli Storage Manager for Windows Version Installation Guide IBM IBM Tivoli Storage Manager for Windows Version 7.1.8 Installation Guide IBM IBM Tivoli Storage Manager for Windows Version 7.1.8 Installation Guide IBM Note: Before you use this information and the product

More information

Presentation Abstract

Presentation Abstract Presentation Abstract From the beginning of DB2, application performance has always been a key concern. There will always be more developers than DBAs, and even as hardware cost go down, people costs have

More information

Oracle Database 10g The Self-Managing Database

Oracle Database 10g The Self-Managing Database Oracle Database 10g The Self-Managing Database Benoit Dageville Oracle Corporation benoit.dageville@oracle.com Page 1 1 Agenda Oracle10g: Oracle s first generation of self-managing database Oracle s Approach

More information

Optimizing Insert Performance - Part 1

Optimizing Insert Performance - Part 1 Optimizing Insert Performance - Part 1 John Campbell Distinguished Engineer DB2 for z/os development CAMPBELJ@uk.ibm.com 2 Disclaimer/Trademarks The information contained in this document has not been

More information

Lock Tuning. Concurrency Control Goals. Trade-off between correctness and performance. Correctness goals. Performance goals.

Lock Tuning. Concurrency Control Goals. Trade-off between correctness and performance. Correctness goals. Performance goals. Lock Tuning Concurrency Control Goals Performance goals Reduce blocking One transaction waits for another to release its locks Avoid deadlocks Transactions are waiting for each other to release their locks

More information

IBM DB2 courses, Universidad Cenfotec

IBM DB2 courses, Universidad Cenfotec IBM DB2 courses, Universidad Cenfotec Contents Summary... 2 Database Management (Information Management) course plan... 3 DB2 SQL Workshop... 4 DB2 SQL Workshop for Experienced Users... 5 DB2 9 Database

More information

Performance Tuning Batch Cycles

Performance Tuning Batch Cycles Performance Tuning Batch Cycles and Table Partitioning Robert Williams One-point Solutions Inc. rwilliams@one-point.com March 11 2010 Platform: LUW Agenda Introduction A little about me About the DWP/EDW

More information

Using Oracle STATSPACK to assist with Application Performance Tuning

Using Oracle STATSPACK to assist with Application Performance Tuning Using Oracle STATSPACK to assist with Application Performance Tuning Scenario You are experiencing periodic performance problems with an application that uses a back-end Oracle database. Solution Introduction

More information

Anthony AWR report INTERPRETATION PART I

Anthony AWR report INTERPRETATION PART I Anthony AWR report INTERPRETATION PART I What is AWR? AWR stands for Automatically workload repository, Though there could be many types of database performance issues, but when whole database is slow,

More information

OASIS: Self-tuning Storage for Applications

OASIS: Self-tuning Storage for Applications OASIS: Self-tuning Storage for Applications Kostas Magoutis, Prasenjit Sarkar, Gauri Shah 14 th NASA Goddard- 23 rd IEEE Mass Storage Systems Technologies, College Park, MD, May 17, 2006 Outline Motivation

More information

Arrays are a very commonly used programming language construct, but have limited support within relational databases. Although an XML document or

Arrays are a very commonly used programming language construct, but have limited support within relational databases. Although an XML document or Performance problems come in many flavors, with many different causes and many different solutions. I've run into a number of these that I have not seen written about or presented elsewhere and I want

More information

MySQL Database Scalability

MySQL Database Scalability MySQL Database Scalability Nextcloud Conference 2016 TU Berlin Oli Sennhauser Senior MySQL Consultant at FromDual GmbH oli.sennhauser@fromdual.com 1 / 14 About FromDual GmbH Support Consulting remote-dba

More information

ORACLE 8 OBJECT ORIENTED TECHNOLOGY ORACLE SOFTWARE STRUCTURES SERVER SIDE BACKGROUND PROCESSES DATABASE SERVER AND DATABASE INSTANCE

ORACLE 8 OBJECT ORIENTED TECHNOLOGY ORACLE SOFTWARE STRUCTURES SERVER SIDE BACKGROUND PROCESSES DATABASE SERVER AND DATABASE INSTANCE ORACLE 8 IS ORDBMS HANDLES VLDB - GIGABYTES/TERABYTES 10,000 CONCURRENT USERS PARTITIONED TABLES AND INDEXES SINGLE DATA BLOCK IS INACCESSSIBLE CAN ACCESS MULTIPLE PARTITION IN PARALLEL FOR CONCURRENT

More information

EMC DMX Disk Arrays with IBM DB2 Universal Database Applied Technology

EMC DMX Disk Arrays with IBM DB2 Universal Database Applied Technology EMC DMX Disk Arrays with IBM DB2 Universal Database Applied Technology Abstract This paper examines the attributes of the IBM DB2 UDB V8.2 database as they relate to optimizing the configuration for the

More information

DB2 is a complex system, with a major impact upon your processing environment. There are substantial performance and instrumentation changes in

DB2 is a complex system, with a major impact upon your processing environment. There are substantial performance and instrumentation changes in DB2 is a complex system, with a major impact upon your processing environment. There are substantial performance and instrumentation changes in versions 8 and 9. that must be used to measure, evaluate,

More information

Performance, Power, Die Yield. CS301 Prof Szajda

Performance, Power, Die Yield. CS301 Prof Szajda Performance, Power, Die Yield CS301 Prof Szajda Administrative HW #1 assigned w Due Wednesday, 9/3 at 5:00 pm Performance Metrics (How do we compare two machines?) What to Measure? Which airplane has the

More information

Advanced Topics UNIT 2 PERFORMANCE EVALUATIONS

Advanced Topics UNIT 2 PERFORMANCE EVALUATIONS Advanced Topics UNIT 2 PERFORMANCE EVALUATIONS Structure Page Nos. 2.0 Introduction 4 2. Objectives 5 2.2 Metrics for Performance Evaluation 5 2.2. Running Time 2.2.2 Speed Up 2.2.3 Efficiency 2.3 Factors

More information

Deep Dive Into Storage Optimization When And How To Use Adaptive Compression. Thomas Fanghaenel IBM Bill Minor IBM

Deep Dive Into Storage Optimization When And How To Use Adaptive Compression. Thomas Fanghaenel IBM Bill Minor IBM Deep Dive Into Storage Optimization When And How To Use Adaptive Compression Thomas Fanghaenel IBM Bill Minor IBM Agenda Recap: Compression in DB2 9 for Linux, Unix and Windows New in DB2 10 for Linux,

More information

Concurrency Control Goals

Concurrency Control Goals Lock Tuning Concurrency Control Goals Concurrency Control Goals Correctness goals Serializability: each transaction appears to execute in isolation The programmer ensures that serial execution is correct.

More information

On BigFix Performance: Disk is King. How to get your infrastructure right the first time! Case Study: IBM Cloud Development - WW IT Services

On BigFix Performance: Disk is King. How to get your infrastructure right the first time! Case Study: IBM Cloud Development - WW IT Services On BigFix Performance: Disk is King How to get your infrastructure right the first time! Case Study: IBM Cloud Development - WW IT Services Authors: Shaun T. Kelley, Mark Leitch Abstract: Rolling out large

More information

Performance of relational database management

Performance of relational database management Building a 3-D DRAM Architecture for Optimum Cost/Performance By Gene Bowles and Duke Lambert As systems increase in performance and power, magnetic disk storage speeds have lagged behind. But using solidstate

More information

Workload Insights Without a Trace - Introducing DB2 z/os SQL tracking SOFTWARE ENGINEERING GMBH and SEGUS Inc. 1

Workload Insights Without a Trace - Introducing DB2 z/os SQL tracking SOFTWARE ENGINEERING GMBH and SEGUS Inc. 1 Workload Insights Without a Trace - Introducing DB2 z/os SQL tracking 2011 SOFTWARE ENGINEERING GMBH and SEGUS Inc. 1 Agenda What s new in DB2 10 What s of interest for geeks in DB2 10 What s of interest

More information

IBM DB2 11 DBA for z/os Certification Review Guide Exam 312

IBM DB2 11 DBA for z/os Certification Review Guide Exam 312 Introduction IBM DB2 11 DBA for z/os Certification Review Guide Exam 312 The purpose of this book is to assist you with preparing for the IBM DB2 11 DBA for z/os exam (Exam 312), one of the two required

More information

IBM DB DBA for LUW Upgrade from DB2 10.1

IBM DB DBA for LUW Upgrade from DB2 10.1 IBM DB2 10.5 DBA for LUW Upgrade from DB2 10.1 Dumps Available Here at: /ibm-exam/c2090-311-dumps.html Enrolling now you will get access to 30 questions in a unique set of C2090-311 dumps Question 1 An

More information

Design of Parallel Algorithms. Course Introduction

Design of Parallel Algorithms. Course Introduction + Design of Parallel Algorithms Course Introduction + CSE 4163/6163 Parallel Algorithm Analysis & Design! Course Web Site: http://www.cse.msstate.edu/~luke/courses/fl17/cse4163! Instructor: Ed Luke! Office:

More information

VMware vrealize operations Management Pack FOR. PostgreSQL. User Guide

VMware vrealize operations Management Pack FOR. PostgreSQL. User Guide VMware vrealize operations Management Pack FOR PostgreSQL User Guide TABLE OF CONTENTS 1. Purpose... 3 2. Introduction to the Management Pack... 3 2.1 How the Management Pack Collects Data... 3 2.2 Data

More information

The Role of Database Aware Flash Technologies in Accelerating Mission- Critical Databases

The Role of Database Aware Flash Technologies in Accelerating Mission- Critical Databases The Role of Database Aware Flash Technologies in Accelerating Mission- Critical Databases Gurmeet Goindi Principal Product Manager Oracle Flash Memory Summit 2013 Santa Clara, CA 1 Agenda Relational Database

More information

Short Summary of DB2 V4 Through V6 Changes

Short Summary of DB2 V4 Through V6 Changes IN THIS CHAPTER DB2 Version 6 Features DB2 Version 5 Features DB2 Version 4 Features Short Summary of DB2 V4 Through V6 Changes This appendix provides short checklists of features for the most recent versions

More information

DB2 10 Capturing Tuning and Trending for SQL Workloads - a resource and cost saving approach

DB2 10 Capturing Tuning and Trending for SQL Workloads - a resource and cost saving approach DB2 10 Capturing Tuning and Trending for SQL Workloads - a resource and cost saving approach Roy Boxwell SOFTWARE ENGINEERING GmbH Session Code: V05 15.10.2013, 11:30 12:30 Platform: DB2 z/os 2 Agenda

More information

1-2 Copyright Ó Oracle Corporation, All rights reserved.

1-2 Copyright Ó Oracle Corporation, All rights reserved. 1-1 The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any

More information

Session: G03 No Magic to Improve DB2 for z/os Application Performance. Marcel Lévy Natixis. May 19, :30 p.m. 02:30 p.m. Platform: DB2 for z/os

Session: G03 No Magic to Improve DB2 for z/os Application Performance. Marcel Lévy Natixis. May 19, :30 p.m. 02:30 p.m. Platform: DB2 for z/os Session: G03 No Magic to Improve DB2 for z/os Application Performance Marcel Lévy Natixis May 19, 2008 01:30 p.m. 02:30 p.m. Platform: DB2 for z/os 1 Agenda Story of a DB2 application migration Measurement

More information

Efficient Bulk Deletes for Multi Dimensional Clustered Tables in DB2

Efficient Bulk Deletes for Multi Dimensional Clustered Tables in DB2 Efficient Bulk Deletes for Multi Dimensional Clustered Tables in DB2 Bishwaranjan Bhattacharjee, Timothy Malkemus IBM T.J. Watson Research Center Sherman Lau, Sean McKeough, Jo-anne Kirton Robin Von Boeschoten,

More information

Rdb features for high performance application

Rdb features for high performance application Rdb features for high performance application Philippe Vigier Oracle New England Development Center Copyright 2001, 2003 Oracle Corporation Oracle Rdb Buffer Management 1 Use Global Buffers Use Fast Commit

More information

Fit for Purpose Platform Positioning and Performance Architecture

Fit for Purpose Platform Positioning and Performance Architecture Fit for Purpose Platform Positioning and Performance Architecture Joe Temple IBM Monday, February 4, 11AM-12PM Session Number 12927 Insert Custom Session QR if Desired. Fit for Purpose Categorized Workload

More information

IBM Optim Performance Manager Extended Edition What s New. Ute Baumbach September 6, IBM Corporation

IBM Optim Performance Manager Extended Edition What s New. Ute Baumbach September 6, IBM Corporation IBM Optim Performance Manager Extended Edition 4.1.1 What s New Ute Baumbach (bmb@de.ibm.com) September 6, 2011 What s New in 4.1.1 Enhancements Summary September 6, 2011 Optim Performance Manager 4.1.1

More information

System Characteristics

System Characteristics System Characteristics Performance is influenced by characteristics of the system hosting the database server, for example: - Disk input/output (I/O) speed. - Amount of memory available. - Processor speed.

More information

To REORG or not to REORG That is the Question. Kevin Baker BMC Software

To REORG or not to REORG That is the Question. Kevin Baker BMC Software To REORG or not to REORG That is the Question Kevin Baker BMC Software Objectives Identify I/O performance trends for DB pagesets Correlate reorganization benefits to I/O performance trends Understand

More information

What it does not show is how to write the program to retrieve this data.

What it does not show is how to write the program to retrieve this data. Session: A16 IFI DATA: IFI you don t know, ask! Jeff Gross CA, Inc. 16 October 2008 11:45 12:45 Platform: DB2 for z/os Abstract The Instrumentation Facility Interface (IFI) can be a daunting resource in

More information

Achieve Breakthrough Performance and Availability with DB2 purescale

Achieve Breakthrough Performance and Availability with DB2 purescale Achieve Breakthrough Performance and Availability with DB2 purescale Philip K. Gunning Gunning Technology Solutions, LLC Session Code: D01 May 3, 2011 12:45 1:45 PM Platform: LUW Building the case for

More information

Introduction to parallel Computing

Introduction to parallel Computing Introduction to parallel Computing VI-SEEM Training Paschalis Paschalis Korosoglou Korosoglou (pkoro@.gr) (pkoro@.gr) Outline Serial vs Parallel programming Hardware trends Why HPC matters HPC Concepts

More information

Optimizing Database I/O

Optimizing Database I/O High Performance Oracle Optimizing Database I/O Dave Pearson Quest Software Copyright 2006 Quest Software The Impact of Poor Performance Diagnostics and Optimization The Impact of Poor Performance Diagnostics

More information

Copyright 2018, Oracle and/or its affiliates. All rights reserved.

Copyright 2018, Oracle and/or its affiliates. All rights reserved. Beyond SQL Tuning: Insider's Guide to Maximizing SQL Performance Monday, Oct 22 10:30 a.m. - 11:15 a.m. Marriott Marquis (Golden Gate Level) - Golden Gate A Ashish Agrawal Group Product Manager Oracle

More information

Range Partitioning Fundamentals

Range Partitioning Fundamentals Session: E08 Range Partitioning Fundamentals Kevin Beck IBM May 8, 2007 4:20 p.m. 5:20 p.m. Platform: DB2 for Linux, UNIX, Windows 1 Agenda Introduction Defining Partitioned Tables Roll-Out Performance

More information

OPS-23: OpenEdge Performance Basics

OPS-23: OpenEdge Performance Basics OPS-23: OpenEdge Performance Basics White Star Software adam@wss.com Agenda Goals of performance tuning Operating system setup OpenEdge setup Setting OpenEdge parameters Tuning APWs OpenEdge utilities

More information

The DB2Night Show Episode #89. InfoSphere Warehouse V10 Performance Enhancements

The DB2Night Show Episode #89. InfoSphere Warehouse V10 Performance Enhancements The DB2Night Show Episode #89 InfoSphere Warehouse V10 Performance Enhancements Pat Bates, WW Technical Sales for Big Data and Warehousing, jpbates@us.ibm.com June 27, 2012 June 27, 2012 Multi-Core Parallelism

More information

Storage Optimization with Oracle Database 11g

Storage Optimization with Oracle Database 11g Storage Optimization with Oracle Database 11g Terabytes of Data Reduce Storage Costs by Factor of 10x Data Growth Continues to Outpace Budget Growth Rate of Database Growth 1000 800 600 400 200 1998 2000

More information

Session: Oracle RAC vs DB2 LUW purescale. Udo Brede Quest Software. 22 nd November :30 Platform: DB2 LUW

Session: Oracle RAC vs DB2 LUW purescale. Udo Brede Quest Software. 22 nd November :30 Platform: DB2 LUW Session: Oracle RAC vs DB2 LUW purescale Udo Brede Quest Software 22 nd November 2011 10:30 Platform: DB2 LUW 1 Agenda Marketing Message Clustering/Scalability Technology Overview Basic Components Available

More information

Jyotheswar Kuricheti

Jyotheswar Kuricheti Jyotheswar Kuricheti 1 Agenda: 1. Performance Tuning Overview 2. Identify Bottlenecks 3. Optimizing at different levels : Target Source Mapping Session System 2 3 Performance Tuning Overview: 4 What is

More information

Basi di Dati Complementi. Mainframe

Basi di Dati Complementi. Mainframe Basi di Dati Complementi 3.1. DBMS commerciali DB2-3.1.2 Db2 in ambiente mainframe Andrea Maurino 2007 2008 Mainframe 1 Mainframe Terminologia Mainframe Storage Management Subsystem (SMS) Is an automated

More information

OpenVMS Performance Update

OpenVMS Performance Update OpenVMS Performance Update Gregory Jordan Hewlett-Packard 2007 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Agenda System Performance Tests

More information

10/29/2013. Program Agenda. The Database Trifecta: Simplified Management, Less Capacity, Better Performance

10/29/2013. Program Agenda. The Database Trifecta: Simplified Management, Less Capacity, Better Performance Program Agenda The Database Trifecta: Simplified Management, Less Capacity, Better Performance Data Growth and Complexity Hybrid Columnar Compression Case Study & Real-World Experiences

More information

Course Outline. Performance Tuning and Optimizing SQL Databases Course 10987B: 4 days Instructor Led

Course Outline. Performance Tuning and Optimizing SQL Databases Course 10987B: 4 days Instructor Led Performance Tuning and Optimizing SQL Databases Course 10987B: 4 days Instructor Led About this course This four-day instructor-led course provides students who manage and maintain SQL Server databases

More information

IBM EXAM - C DB DBA for Linux, UNIX, and Windows. Buy Full Product.

IBM EXAM - C DB DBA for Linux, UNIX, and Windows. Buy Full Product. IBM EXAM - C2090-611 DB2 10.1 DBA for Linux, UNIX, and Windows Buy Full Product http://www.examskey.com/c2090-611.html Examskey IBM C2090-611 exam demo product is here for you to test the quality of the

More information

Final Lecture. A few minutes to wrap up and add some perspective

Final Lecture. A few minutes to wrap up and add some perspective Final Lecture A few minutes to wrap up and add some perspective 1 2 Instant replay The quarter was split into roughly three parts and a coda. The 1st part covered instruction set architectures the connection

More information

Computer Architecture. R. Poss

Computer Architecture. R. Poss Computer Architecture R. Poss 1 ca01-10 september 2015 Course & organization 2 ca01-10 september 2015 Aims of this course The aims of this course are: to highlight current trends to introduce the notion

More information

IFCID Instrumentation Accounting Data This topic shows detailed information about Record Trace - IFCID Instrumentation Accounting Data.

IFCID Instrumentation Accounting Data This topic shows detailed information about Record Trace - IFCID Instrumentation Accounting Data. This topic shows detailed information about Record Trace - IFCID 003 - Instrumentation Accounting Data. Note: IFCID 003 and IFCID 147 have the same layout. IFCID 003 - Instrumentation Accounting Data Record

More information

Optimizing RDM Server Performance

Optimizing RDM Server Performance TECHNICAL WHITE PAPER Optimizing RDM Server Performance A Raima Inc. Technical Whitepaper Published: August, 2008 Author: Paul Johnson Director of Marketing Copyright: Raima Inc., All rights reserved Abstract

More information

CSE 544: Principles of Database Systems

CSE 544: Principles of Database Systems CSE 544: Principles of Database Systems Anatomy of a DBMS, Parallel Databases 1 Announcements Lecture on Thursday, May 2nd: Moved to 9am-10:30am, CSE 403 Paper reviews: Anatomy paper was due yesterday;

More information

An Analysis of Linux Scalability to Many Cores

An Analysis of Linux Scalability to Many Cores An Analysis of Linux Scalability to Many Cores 1 What are we going to talk about? Scalability analysis of 7 system applications running on Linux on a 48 core computer Exim, memcached, Apache, PostgreSQL,

More information

B.H.GARDI COLLEGE OF ENGINEERING & TECHNOLOGY (MCA Dept.) Parallel Database Database Management System - 2

B.H.GARDI COLLEGE OF ENGINEERING & TECHNOLOGY (MCA Dept.) Parallel Database Database Management System - 2 Introduction :- Today single CPU based architecture is not capable enough for the modern database that are required to handle more demanding and complex requirements of the users, for example, high performance,

More information

OS and Hardware Tuning

OS and Hardware Tuning OS and Hardware Tuning Tuning Considerations OS Threads Thread Switching Priorities Virtual Memory DB buffer size File System Disk layout and access Hardware Storage subsystem Configuring the disk array

More information

Pass IBM C Exam

Pass IBM C Exam Pass IBM C2090-612 Exam Number: C2090-612 Passing Score: 800 Time Limit: 120 min File Version: 37.4 http://www.gratisexam.com/ Exam Code: C2090-612 Exam Name: DB2 10 DBA for z/os Certkey QUESTION 1 Workload

More information

The former pager tasks have been replaced in 7.9 by the special savepoint tasks.

The former pager tasks have been replaced in 7.9 by the special savepoint tasks. 1 2 3 4 With version 7.7 the I/O interface to the operating system has been reimplemented. As of version 7.7 different parameters than in version 7.6 are used. The improved I/O system has the following

More information

Multidimensional Clustering (MDC) Tables in DB2 LUW. DB2Night Show. January 14, 2011

Multidimensional Clustering (MDC) Tables in DB2 LUW. DB2Night Show. January 14, 2011 Multidimensional Clustering (MDC) Tables in DB2 LUW DB2Night Show January 14, 2011 Pat Bates, IBM Technical Sales Professional, Data Warehousing Paul Zikopoulos, Director, IBM Information Management Client

More information

OS and HW Tuning Considerations!

OS and HW Tuning Considerations! Administração e Optimização de Bases de Dados 2012/2013 Hardware and OS Tuning Bruno Martins DEI@Técnico e DMIR@INESC-ID OS and HW Tuning Considerations OS " Threads Thread Switching Priorities " Virtual

More information

Bipul Sinha, Amit Ganesh, Lilian Hobbs, Oracle Corp. Dingbo Zhou, Basavaraj Hubli, Manohar Malayanur, Fannie Mae

Bipul Sinha, Amit Ganesh, Lilian Hobbs, Oracle Corp. Dingbo Zhou, Basavaraj Hubli, Manohar Malayanur, Fannie Mae ONE MILLION FINANCIAL TRANSACTIONS PER HOUR USING ORACLE DATABASE 10G AND XA Bipul Sinha, Amit Ganesh, Lilian Hobbs, Oracle Corp. Dingbo Zhou, Basavaraj Hubli, Manohar Malayanur, Fannie Mae INTRODUCTION

More information

Bottleneck Hunters: How Schooner increased MySQL throughput by more than 800% Jeremy Cole

Bottleneck Hunters: How Schooner increased MySQL throughput by more than 800% Jeremy Cole Bottleneck Hunters: How Schooner increased MySQL throughput by more than 800% Jeremy Cole On the genesis of Schooner: Hardware is massively under-utilized I/O has long

More information

Estimate performance and capacity requirements for Access Services

Estimate performance and capacity requirements for Access Services Estimate performance and capacity requirements for Access Services This document is provided as-is. Information and views expressed in this document, including URL and other Internet Web site references,

More information

Lesson 4 Transcript: DB2 Architecture

Lesson 4 Transcript: DB2 Architecture Lesson 4 Transcript: DB2 Architecture Slide 1: Cover Welcome to Lesson 4 of the DB2 on campus series. Today we are going to talk about the DB2 architecture. My name is Raul Chong and I am the DB2 on Campus

More information

Field Testing Buffer Pool Extension and In-Memory OLTP Features in SQL Server 2014

Field Testing Buffer Pool Extension and In-Memory OLTP Features in SQL Server 2014 Field Testing Buffer Pool Extension and In-Memory OLTP Features in SQL Server 2014 Rick Heiges, SQL MVP Sr Solutions Architect Scalability Experts Ross LoForte - SQL Technology Architect - Microsoft Changing

More information

DB2 UDB: App Programming - Advanced

DB2 UDB: App Programming - Advanced A Access Methods... 8:6 Access Path Selection... 8:6 Access Paths... 5:22 ACQUIRE(ALLOCATE) / RELEASE(DEALLOCATE)... 5:14 ACQUIRE(USE) / RELEASE(DEALLOCATE)... 5:14 Active Log... 9:3 Active Logs - Determining

More information

Page 1. Program Performance Metrics. Program Performance Metrics. Amdahl s Law. 1 seq seq 1

Page 1. Program Performance Metrics. Program Performance Metrics. Amdahl s Law. 1 seq seq 1 Program Performance Metrics The parallel run time (Tpar) is the time from the moment when computation starts to the moment when the last processor finished his execution The speedup (S) is defined as the

More information

Evaluating SMB2 Performance for Home Directory Workloads

Evaluating SMB2 Performance for Home Directory Workloads Evaluating SMB2 Performance for Home Directory Workloads Dan Lovinger, David Kruse Development Leads Windows Server / File Server Team 2010 Storage Developer Conference. Microsoft Corporation. All Rights

More information

!! What is virtual memory and when is it useful? !! What is demand paging? !! When should pages in memory be replaced?

!! What is virtual memory and when is it useful? !! What is demand paging? !! When should pages in memory be replaced? Chapter 10: Virtual Memory Questions? CSCI [4 6] 730 Operating Systems Virtual Memory!! What is virtual memory and when is it useful?!! What is demand paging?!! When should pages in memory be replaced?!!

More information

Extracting Performance and Scalability Metrics from TCP. Baron Schwartz April 2012

Extracting Performance and Scalability Metrics from TCP. Baron Schwartz April 2012 Extracting Performance and Scalability Metrics from TCP Baron Schwartz April 2012 Agenda Fundamental Metrics of Performance Capturing TCP Data Part 1: Black-Box Performance Analysis Detecting Stalls and

More information