How Much Memory Do I Need? Jack Opgenorth October, 2004.

Slides:



Advertisements
Similar presentations
3D Graphics Content Over OCP Martti Venell Sr. Verification Engineer Bitboys.
Advertisements

Accessing I/O Devices Processor Memory BUS I/O Device 1 I/O Device 2.
Paper by: Yu Li, Jianliang Xu, Byron Choi, and Haibo Hu Department of Computer Science Hong Kong Baptist University Slides and Presentation By: Justin.
CSE 490/590, Spring 2011 CSE 490/590 Computer Architecture Cache III Steve Ko Computer Sciences and Engineering University at Buffalo.
Operating Systems Lecture 10 Issues in Paging and Virtual Memory Adapted from Operating Systems Lecture Notes, Copyright 1997 Martin C. Rinard. Zhiqing.
Kernel memory allocation
1 Hash-Based Indexes Module 4, Lecture 3. 2 Introduction As for any index, 3 alternatives for data entries k* : – Data record with key value k – –Choice.
Hash-Based Indexes The slides for this text are organized into chapters. This lecture covers Chapter 10. Chapter 1: Introduction to Database Systems Chapter.
Index tuning Hash Index. overview Introduction Hash-based indexes are best for equality selections. –Can efficiently support index nested joins –Cannot.
G Robert Grimm New York University Sprite LFS or Let’s Log Everything.
File Management Systems
1 CSSE 477 – A bit more on Performance Steve Chenoweth Friday, 9/9/11 Week 1, Day 2 Right – Googling for “Performance” gets you everything from Lady Gaga.
Operating System Support Focus on Architecture
I/O Hardware n Incredible variety of I/O devices n Common concepts: – Port – connection point to the computer – Bus (daisy chain or shared direct access)
OS Fall ’ 02 Performance Evaluation Operating Systems Fall 2002.
1 Hash-Based Indexes Chapter Introduction  Hash-based indexes are best for equality selections. Cannot support range searches.  Static and dynamic.
METU Department of Computer Eng Ceng 302 Introduction to DBMS Disk Storage, Basic File Structures, and Hashing by Pinar Senkul resources: mostly froom.
Performance Evaluation
SECTIONS 13.1 – 13.3 Sanuja Dabade & Eilbroun Benjamin CS 257 – Dr. TY Lin SECONDARY STORAGE MANAGEMENT.
Copyright © 2007 Ramez Elmasri and Shamkant B. Navathe Chapter 13 Disk Storage, Basic File Structures, and Hashing.
1 COMP 206: Computer Architecture and Implementation Montek Singh Wed, Nov 9, 2005 Topic: Caches (contd.)
1 CS 501 Spring 2005 CS 501: Software Engineering Lecture 22 Performance of Computer Systems.
Implications for Programming Models Todd C. Mowry CS 495 September 12, 2002.
E.G.M. PetrakisHashing1 Hashing on the Disk  Keys are stored in “disk pages” (“buckets”)  several records fit within one page  Retrieval:  find address.
OS Fall ’ 02 Performance Evaluation Operating Systems Fall 2002.
1 Exploring Data Reliability Tradeoffs in Replicated Storage Systems NetSysLab The University of British Columbia Abdullah Gharaibeh Matei Ripeanu.
SECTIONS 13.1 – 13.3 Sanuja Dabade & Eilbroun Benjamin CS 257 – Dr. TY Lin SECONDARY STORAGE MANAGEMENT.
Performance Management (Best Practices) REF: Document ID
Page 1 © 2001 Hewlett-Packard Company Tools for Measuring System and Application Performance Introduction GlancePlus Introduction Glance Motif Glance Character.
THE DESIGN AND IMPLEMENTATION OF A LOG-STRUCTURED FILE SYSTEM M. Rosenblum and J. K. Ousterhout University of California, Berkeley.
Chapter 9 Overview  Reasons to monitor SQL Server  Performance Monitoring and Tuning  Tools for Monitoring SQL Server  Common Monitoring and Tuning.
Basics of Operating Systems March 4, 2001 Adapted from Operating Systems Lecture Notes, Copyright 1997 Martin C. Rinard.
From Here to There Thiru Thirupuvanam, Ed Jordan & Jeff Longwell HERE THERE.
1 © Copyright 2008 EMC Corporation. All rights reserved. Symmetrix Capacity Planning and Performance Aspects Bob Rau Technical Business Consultant Symmetrix.
Toolbox for Dimensioning Windows Storage Systems Jalil Boukhobza, Claude Timsit 12/09/2006 Versailles Saint Quentin University.
Copyright © 2011 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Chapter 17 Disk Storage, Basic File Structures, and Hashing.
Parallel Programming Models Jihad El-Sana These slides are based on the book: Introduction to Parallel Computing, Blaise Barney, Lawrence Livermore National.
Computer Architecture and Operating Systems CS 3230: Operating System Section Lecture OS-7 Memory Management (1) Department of Computer Science and Software.
Enterprise PI - How do I manage all of this? Robert Raesemann J Jacksonville, FL.
Recall: Three I/O Methods Synchronous: Wait for I/O operation to complete. Asynchronous: Post I/O request and switch to other work. DMA (Direct Memory.
Suite zTPFGI Facilities. Suite Focus Three of zTPFGI’s facilities:  zAutomation  zTREX  Logger.
Windows 2000 Course Summary Computing Department, Lancaster University, UK.
A Measurement Based Memory Performance Evaluation of High Throughput Servers Garba Isa Yau Department of Computer Engineering King Fahd University of Petroleum.
Suite zTPFGI Facilities. Suite Focus Three of zTPFGI’s facilities:  zAutomation  zTREX  Logger.
Performance evaluation of component-based software systems Seminar of Component Engineering course Rofideh hadighi 7 Jan 2010.
Operating Systems. Overview What is an Operating System (OS) What is an Operating System (OS) What Operating Systems do. What Operating Systems do. Operating.
Accessing I/O Devices Processor Memory BUS I/O Device 1 I/O Device 2.
Paper Survey of DHT Distributed Hash Table. Usages Directory service  Very little amount of information, such as URI, metadata, … Storage  Data, such.
Lecture 3 Page 1 CS 111 Online Disk Drives An especially important and complex form of I/O device Still the primary method of providing stable storage.
File Structures. 2 Chapter - Objectives Disk Storage Devices Files of Records Operations on Files Unordered Files Ordered Files Hashed Files Dynamic and.
Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke1 Indexed Sequential Access Method.
Database Management Systems, R. Ramakrishnan and J. Gehrke1 Hash-Based Indexes Chapter 10.
Introduction to z/OS Basics © 2006 IBM Corporation Chapter 7: Batch processing and the Job Entry Subsystem (JES) Batch processing and JES.
Storage Systems CSE 598d, Spring 2007 Lecture ?: Rules of thumb in data engineering Paper by Jim Gray and Prashant Shenoy Feb 15, 2007.
Chapter 5 Record Storage and Primary File Organizations
BNL dCache Status and Plan CHEP07: September 2-7, 2007 Zhenping (Jane) Liu for the BNL RACF Storage Group.
Software System Performance CS 560. Performance of computer systems In most computer systems:  The cost of people (development) is much greater than.
If you have a transaction processing system, John Meisenbacher
Credit and Collections optimization and decision support An example of the intersection of decision tool development and management consulting Includes.
System Components Operating System Services System Calls.
Modeling the Optimization of a Toll Booth Plaza By Liam Connell, Erin Hoover and Zach Schutzman Figure 1. Plot of k values and outputs from Erlang’s Formula.
Traffic Simulation L2 – Introduction to simulation Ing. Ondřej Přibyl, Ph.D.
Memory Management for Scalable Web Data Servers
Sanuja Dabade & Eilbroun Benjamin CS 257 – Dr. TY Lin
Chapter 9: Virtual-Memory Management
Intro. To Operating Systems
KISS-Tree: Smart Latch-Free In-Memory Indexing on Modern Architectures
Overview of Computer Architecture and Organization
Overview of Computer Architecture and Organization
Introduction to Operating Systems
Presentation transcript:

How Much Memory Do I Need? Jack Opgenorth October, 2004

Page 2 Overview z/TPF = Relief, –But How Soon Can You Get There? How Soon Will I Run Out? What Are the Driving Factors? If Necessary, Can You Extend Your Timeline?

Page 3 Z is for Zillions of Bytes Trivia Challenge What’s 2 64 /2 31 GB?  A HEAP! z/TPF 1.1 –Lots of Storage –Lots of other Features, too Do You Want To β Your Business?

Page 4 Running on Empty? Some of Us May Already Know We Have A Problem How Are You Going to Know? –Measure (Last Spring’s Presentation) –Evaluate the Data Get The Brick Stretcher Out! Change Parameters  $ Change Applications $$$

Page 5 TPF Storage Distribution

Page 6 Drivers of Storage Growth Globals –Increased complexity of data market segmentation-creates replication –Volume of Data in Memory Continues to Grow Working Storage –Heap Storage – MBs / ECB –More Engines (16 way) –Larger and more complex data structures –Increased Latency (IO)

Page 7 What Drives Working Storage Demand Application Use of Heap Is One Contributor But Existence Time Is A Significant Contributor N = Φ*R (Little’s Law) Simplified  Working Storage Required (N) = Existence Time (Φ) * Arrival Rate (R) All Values Can Be Obtained From Data Collection Generally, Φ And R Together

Page 8 Are Your Cheeks Full? You’ve Probably Seen The Phone Ads –It’s “Dang-shin, Voce, ‘Nee’, Sie, Usted, You …..” High Utilizations

Page 9 Another Example

Page 10 Another Example- Is It All IO?

Page 11 What Else Is There? What Are We Waiting On? –IO or Someone Else That Looks Like An IO Distributed Processing-Synchronous Calls? Do You Have Time-Out Thresholds For Certain Traffic? Scenario 1 vs. Scenario 2 44% Increase In Existence Time Results (Ouch!) Traffic TypeMsg. Rate Existence Time (sec) Internal Only 10, Distributed Total 10, Traffic TypeMsg. Rate Existence Time (sec) Internal Only 10, Distributed Total 10,

Page 12 Possible Solutions Tighten Up Response Time Requirements –How Many Messages With 10 Sec T.O. Are You Holding Onto? Internal Queuing IO Latency I-Stream Unique 31 Bit Globals? –Increasing I-Streams Reduces VFA –Recall VFA Performance vs. Memory (Blackburn)

Page 13 Some Opportunities for Everyone 1 ST Check Data Collection VFA EFFICIENCY STATISTICS SYSTEM WIDE BSS SUBSYSTEM PROGRAM READS 0.00 PER SECOND 0.00 PER SECOND DATA READS PER SECOND PER SECOND FINDS (WITH I/O) PER SECOND PER SECOND CANDIDATE FILES PER SECOND PER SECOND FORCE FILES 2.70 PER SECOND 2.70 PER SECOND NON-CANDIDATE FILES PER SECOND PER SECOND FILE IMMEDIATES PER SECOND PER SECOND BUFFER UNAVAILABLE 0.00 PER SECOND 0.00 PER SECOND  381 BUFFER USAGE 3.63 PER SECOND 3.63 PER SECOND 1055 BUFFER USAGE PER SECOND PER SECOND 4K BUFFER USAGE PER SECOND PER SECOND ….

Page 14 VFA Buffer Allocation Use Data Available From APAR –File Address –Access Rate Other’s Ideas Used- Aging Time (Residency) Access Density (Spring, 2004) Base Buffer Allocations on Locality Of Reference Don’t Forget About Utilities No Worries in z/TPF

Page 15 Building A Bridge to 64 Bit Architecture Instrumentation is Key –Data Collection Working Storage, CRPA, VFA … No Help for Globals – Approaching 50% of Consumption –Choices for Data in Memory left to Subject Matter Experts Good Choices Can Be Made Creep in Data Structures Diffuse Benefit

Page 16 Global Measurement (Spring, 2004) Tool Samples Memory Access Based on Event Timer (1 second Intervals) Range of Memory Sampled Reports Demonstrate Global Recovery Opportunity

Page 17 Know Your Workload Memory Modeling Is Not Rocket Science –Little’s Law –Apply It –Check With Reality Watch Out For The Other YOU’s CPU Speed Is Not Our Biggest Problem –Know What You Waiting On (IO, Delay, Defer, and Synchronous Traffic

Page 18 Questions ?

eds.com