Computational performance and scalability of large distributed enterprise-wide systems supporting engineering, manufacturing and business applications
|
|
- Edwin Short
- 5 years ago
- Views:
Transcription
1 Computational performance and scalability of large distributed enterprise-wide systems supporting engineering, manufacturing and business applications Janusz S. Kowalik Mathematics and Computing Technology The Boeing Company Abstract The Boeing Company has been implementing large client/server systems that support all major engineering, manufacturing and business functions. One of the challenges has been computational performance and scalability to a very large number of users and data sets. Modeling and simulation combined with analytical approximations have been used to predict performance, identify bottlenecks and to establish the required system capacity. This approach is highly dependent on the accurate workload predictions which usually are uncertain and changing. To get around this difficulty the system models have been used in conjunction with parametric studies that explore ranges of system performance behavior. Good performance and scalability are important system quality indicators but there are other issues that have to be taken into account such as system's cost and availability. One of our recent research projects aims at designing a method for defining an initial system topology and capacity. This definition is not unique and can be further refined by considering system's costs. The three models: workload model, performance model and cost model together allow trade-off studies and making rational decisions about the final system selection in view of the present and the future technological and business uncertainties. 1 Introduction We consider here large client/server distributed transaction systems that are used by tens of thousands of users and run applications such as BAAN, PeopleSoft and
2 392 Applications 2000 WIT Press, of High-Performance ISBN Computers in Engineering VI CATIA (a CAD including : system). A system of this kind must satisfy several requirements (a) computational performance (e.g. response times and thruputs) (b) scalability (allow new users and new applications), (c) to be built and maintained at an acceptable cost. Since such systems are company strategic assets they have to be reliable with a high uptime to support business processes. It has been observed that systems supporting emergent organizations (dynamically evolving and constantly seeking new business modes of operations) cannot be stable and finished. The authors of a recent paper on this subject (1) stated "Systems should be under constant development, can never be fully specified and are subject to constant adjustment and adaptation". We can add that not only organizational needs are changing but also fast moving technology makes any static system increasingly cost/inefficient and eventually obsolete. It is not easy to build a system that is not only cost effective at the time when it is delivered but it evolves in time and has a long life with economic functionality. We, the IS community do not have at present a proven methodology for designing such systems but we have made first steps that have advanced our understanding of the issues, methods and tools that are needed for accomplishing the goal. In this paper we share some of our lessons learned and describe the related experience. 1.1 Technical challenges This paper also attempts to summarize an approach to designing and reengineering large client/server architectures and make specific suggestions for organizations which require such systems. A brief description of technical challenges is a good starting point. (a) Workload predictions are uncertain and inaccurate. Unless the planned system has been at least partly in production and needs only tuning there is no production workload and utilization data available. Vendor benchmark data may not correspond to the expected system use and typically have been obtained using a scaled down system compared to the target system. (b) Multiple vendors environment tends to have ill-defined responsibility for the total system integration, functionality and performance. (c) System size (the number of users and the total system's workload) is often beyond vendors'experience. (d) Vendors are tempted to provide simple linear extrapolations of performance and capacity data from small systems to larger configurations. This may lead to technical and cost disappointments. Correct extrapolations require complex performance models exhibiting highly nonlinear behavior. (e) We define system scalability as the ability to add system resources in order to handle additional workloads exceeding the workloads used in the design definition, without catastrophic performance degradation or unreasonable expenses. System scalability requires scalable hardware, software an communications. To put it differently the useful notion of scalability includes the entire system not any part of it.
3 2000 WIT Press, Applications of High-Performance ISBN Computers in Engineering VI 393 (f) We have to design a system for all applications running concurrently. It is not enough to consider separate applications even if their independent performances are satisfactory. 1.2 Performance and Scalability Factors Total Hardware/Software Fit Workload Characteristics Application Software: Interconnected COTS systems System software: OS, Schedulers, load balancing Hardware: Servers and networking Figure 1: Factors influencing performance Consider the question: what does determine system performance and scalability? The answer to the question is potentially everything. Any system component can become a bottleneck limiting the system performance. A well balance system does not have any resource device too close to saturation (100% utilization). Otherwise this device is likely to become a bottleneck. (a) a well balanced system for one workload may be out-of-balance for another. (b) but even a perfectly balanced system may be performing poorly for other reasons than device saturation. For example: excessive remote memory accesses, bad thread scheduling algorithms which defeat the memory - CPU affinity principle, etc. (c) While designing a system we tend to select high performance components, but this by itself does not warrant high performance for the integrated system (d) Unfortunately, it is also true that a single weak component that is on the transaction paths can degrade the end to end system performance and scalability. Summing up this section we conclude that a thorough capacity planning exercise which determines required resources such as processing power and memory has to be combined with performance studies to avoid pitfalls of having a system with plenty of resources and poor performance at the same time.
4 394 Applications of High-Performance Computers in Engineering VI 2 Approach The approach to designing and tuning large client/server distributed architectures that has proven to be cost-effective is building and exercising two types of models: workload models and performance models. Two kinds of modeling techniques are combined: analytical and simulation. Analytical modeling helps to evaluate performance of single components or simplified systems in order to identify performance concerns and find bottlenecks. Simulation modeling provides more detailed performance characteristics of the integrated global systems. 2.1 Formal problem definition Formally the system performance problem resembles a complex optimization task: given a workload and service level agreements (SLA) design a system topology and capacity that satisfies SLAs subject to cost constrains. The most desirable system may not be least expensive.since the owner may prefer to minimize the future risks (lack of scalability, redesign disruption, or inability to upgrade the system ) rather than minimize the initial system cost. There are two distinct cases which require different approaches: (a) there is already a production system that can be used to obtain data needed for modeling,and (b) there is no production data source,but some benchmarks are available and additional data can be obtained from simplified analytical models approximating the target system. 2.2 Modeling objectives and tools Our main simulation modeling tools have been Workbench and Strategizer from the SES (Scientific Engineering Systems ) Company. Using these tools together with analytical modeling we have been able to perform the required work: (a) capacity planning studies (b) performance evaluation ( response times and thruputs) (c) what-if studies for different architectural and workload assumptions, (d) identify and remove performance bottlenecks The objectives of our work have depended on the system development stage. The most preferable stage is the very beginning of the project before significant architectural decisions are made. At more mature stages of system development only tuning is an option. It is important to realize that performance models should be kept and exercised after the system is implemented. The idea is to modify the system continuously by modernizing its components rather than wait until the system is obsolete and has to be replaced. This modernization process can greatly benefit from the availability of performance models and help to avoid costly replacement and disrupting company business.
5 Applications of High-Performance Computers in Engineering VI Information required for modeling The first step needed for building a simulation performance model is gathering data that characterize the system including the expected workload. This is summarized in Fig 2. We need to know the types and mix of transactions, transaction flow through the system, arrival rates and utilization of resources such as CPU and memory. We may be interested in average performance or performance for the peak loads. The next type of information required is related to hardware and software that we intend to use. This includes the initial topology of the system (how devices are connected), speeds and capacities of servers and their architectures (e.g. uniprocessors or multiprocessors) and other hardware components such as disks and networks. In the software area we need to know some characteristics of the DBMS and operating systems (OS).OS are responsible for the CPU scheduling methods, load balancing and managing concurrent threads.all these factors impact system computational performance. Some new server architectures such as NCC-NUMA (cache coherent non-uniform.memory access) are very sensitive to these software system features. Given this information we can build within a couple of weeks a medium or a large system model. The greatest impediment to modeling is lack of reliable data. Another not surprisingly is the computational performance of the tools themselves. Good performance and scalability of tools is essential to their efficient and successful use. Workload characteristics types of transactions transaction flow arrival rates mix of transactions intensity: number of active users H/W and SAV characteristics Configuration topology Processor speeds/parallelism Disk parameters Network parameters. OS CPU scheduler Concurrent threads Load balancing algorithms Figure 2: Information needed for modeling
6 396 Applications of High-Performance Computers in Engineering VI 3 Examples To illustrate the usefulness of analytical modeling we provide an example of a model of a commercial UNIX SMP server. The following notation is used: m - The number of processors U- CPU utilization S - Service Time 8-U/m R -Response Time Approx. R =, The response time equation is developed using the well known Little relationships. This equation is approximate and does underestimate time. More accurate response time can be calculated from the Erland function (3). Clearly the system response time is a nonlinear function of the number of processors. Adding more processors to the configuration reaches a point of diminishing return (Fig. 3) m Figure 3: Response time as a function of m
7 Applications of High-Performance Computers in Engineering VI 397 Figure 4 shows that for heavy loads close to 1.00 multiprocessor performance suffers more drastically than the performance of the uniprocessor I i I I i Load 4 Conclusions Figure 4: Another View Adequate resource capacity is required for good performance but it does not guarantee it Therefore adding resources may be counterproductive, i.e. may not improve performance unless we understand performance problems and bottlenecks One of the hardest bottlenecks to find and eliminate are those related to software (system and application) References [1] Truex, D.P, Baskerville, R. & Klein, H. Growing systems in emergent organizations, CACM,Vol.42, No. 8, pp ,1999. [2] Menasce, D.A., Almeida, V. A. F. and Dowdy, L. W. Capacity planning and performance modeling, Prentice Hall, [3] Gunther, N. J. The Practical Performance Analyst, The McGraw Hill Series on Computer Communications, 1998.
Future-ready IT Systems with Performance Prediction using Analytical Models
Future-ready IT Systems with Performance Prediction using Analytical Models Madhu Tanikella Infosys Abstract Large and complex distributed software systems can impact overall software cost and risk for
More informationA Quantitative Model for Capacity Estimation of Products
A Quantitative Model for Capacity Estimation of Products RAJESHWARI G., RENUKA S.R. Software Engineering and Technology Laboratories Infosys Technologies Limited Bangalore 560 100 INDIA Abstract: - Sizing
More informationVirtualizing the SAP Infrastructure through Grid Technology. WHITE PAPER March 2007
Virtualizing the SAP Infrastructure through Grid Technology WHITE PAPER March 2007 TABLE OF CONTENTS TABLE OF CONTENTS 2 Introduction 3 The Complexity of the SAP Landscape 3 Specific Pain Areas 4 Virtualizing
More informationPerformance Extrapolation for Load Testing Results of Mixture of Applications
Performance Extrapolation for Load Testing Results of Mixture of Applications Subhasri Duttagupta, Manoj Nambiar Tata Innovation Labs, Performance Engineering Research Center Tata Consulting Services Mumbai,
More informationLecture 9: MIMD Architectures
Lecture 9: MIMD Architectures Introduction and classification Symmetric multiprocessors NUMA architecture Clusters Zebo Peng, IDA, LiTH 1 Introduction A set of general purpose processors is connected together.
More informationMassive Data Analysis
Professor, Department of Electrical and Computer Engineering Tennessee Technological University February 25, 2015 Big Data This talk is based on the report [1]. The growth of big data is changing that
More informationCommentary. EMC VPLEX Launches the Virtual Storage Era
Mesabi Group Commentary May 10, 2010 EMC VPLEX Launches the Virtual Storage Era Magicians make objects appear, disappear, or change appearance. EMC is doing the same thing with information and long-held
More informationTeradata Analyst Pack More Power to Analyze and Tune Your Data Warehouse for Optimal Performance
Data Warehousing > Tools & Utilities Teradata Analyst Pack More Power to Analyze and Tune Your Data Warehouse for Optimal Performance By: Rod Vandervort, Jeff Shelton, and Louis Burger Table of Contents
More informationvsan 6.6 Performance Improvements First Published On: Last Updated On:
vsan 6.6 Performance Improvements First Published On: 07-24-2017 Last Updated On: 07-28-2017 1 Table of Contents 1. Overview 1.1.Executive Summary 1.2.Introduction 2. vsan Testing Configuration and Conditions
More informationNEXGEN N5 PERFORMANCE IN A VIRTUALIZED ENVIRONMENT
NEXGEN N5 PERFORMANCE IN A VIRTUALIZED ENVIRONMENT White Paper: NexGen N5 Performance in a Virtualized Environment January 2015 Contents Introduction... 2 Objective... 2 Audience... 2 NexGen N5... 2 Test
More informationSCALING UP VS. SCALING OUT IN A QLIKVIEW ENVIRONMENT
SCALING UP VS. SCALING OUT IN A QLIKVIEW ENVIRONMENT QlikView Technical Brief February 2012 qlikview.com Introduction When it comes to the enterprise Business Discovery environments, the ability of the
More informationShared Memory and Distributed Multiprocessing. Bhanu Kapoor, Ph.D. The Saylor Foundation
Shared Memory and Distributed Multiprocessing Bhanu Kapoor, Ph.D. The Saylor Foundation 1 Issue with Parallelism Parallel software is the problem Need to get significant performance improvement Otherwise,
More informationPerformance 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 informationLecture 9: MIMD Architectures
Lecture 9: MIMD Architectures Introduction and classification Symmetric multiprocessors NUMA architecture Clusters Zebo Peng, IDA, LiTH 1 Introduction MIMD: a set of general purpose processors is connected
More informationCHAPTER 3: LITERATURE REVIEW
CHAPTER 3: LITERATURE REVIEW 3.1 INTRODUCTION The information in an organization can be categorized in three ways according to need of managerial level i.e. strategic information used by top management
More informationCh. 7: Benchmarks and Performance Tests
Ch. 7: Benchmarks and Performance Tests Kenneth Mitchell School of Computing & Engineering, University of Missouri-Kansas City, Kansas City, MO 64110 Kenneth Mitchell, CS & EE dept., SCE, UMKC p. 1/3 Introduction
More informationModule 5: Performance Issues in Shared Memory and Introduction to Coherence Lecture 10: Introduction to Coherence. The Lecture Contains:
The Lecture Contains: Four Organizations Hierarchical Design Cache Coherence Example What Went Wrong? Definitions Ordering Memory op Bus-based SMP s file:///d /...audhary,%20dr.%20sanjeev%20k%20aggrwal%20&%20dr.%20rajat%20moona/multi-core_architecture/lecture10/10_1.htm[6/14/2012
More informationOptimisation drives digital transformation
January 2017 Executive summary Forward-thinking business leaders are challenging their organisations to achieve transformation by harnessing digital technologies with organisational, operational, and business
More informationSolution Guide. 10 Non-Negotiables of IT Infrastructure Performance Management
Solution Guide 10 Non-Negotiables of IT Infrastructure Performance Management Many IT optimization efforts fail because of management s inability to recognize the importance of an integrated infrastructure
More informationFlorida Board of Governors General Office Legislative Budget Request
Florida Board of Governors General Office 2018-2019 Legislative Budget Request Funding of $9.16 million is needed to support the 65 authorized positions and associated operating expense for the Board Office.
More informationCS 590: High Performance Computing. Parallel Computer Architectures. Lab 1 Starts Today. Already posted on Canvas (under Assignment) Let s look at it
Lab 1 Starts Today Already posted on Canvas (under Assignment) Let s look at it CS 590: High Performance Computing Parallel Computer Architectures Fengguang Song Department of Computer Science IUPUI 1
More informationDELL EMC CX4 EXCHANGE PERFORMANCE THE ADVANTAGES OF DEPLOYING DELL/EMC CX4 STORAGE IN MICROSOFT EXCHANGE ENVIRONMENTS. Dell Inc.
DELL EMC CX4 EXCHANGE PERFORMANCE THE ADVANTAGES OF DEPLOYING DELL/EMC CX4 STORAGE IN MICROSOFT EXCHANGE ENVIRONMENTS Dell Inc. October 2008 Visit www.dell.com/emc for more information on Dell/EMC Storage.
More informationOperating System Performance and Large Servers 1
Operating System Performance and Large Servers 1 Hyuck Yoo and Keng-Tai Ko Sun Microsystems, Inc. Mountain View, CA 94043 Abstract Servers are an essential part of today's computing environments. High
More informationMessage Passing. Advanced Operating Systems Tutorial 5
Message Passing Advanced Operating Systems Tutorial 5 Tutorial Outline Review of Lectured Material Discussion: Barrelfish and multi-kernel systems Programming exercise!2 Review of Lectured Material Implications
More informationSoftNAS Cloud Performance Evaluation on AWS
SoftNAS Cloud Performance Evaluation on AWS October 25, 2016 Contents SoftNAS Cloud Overview... 3 Introduction... 3 Executive Summary... 4 Key Findings for AWS:... 5 Test Methodology... 6 Performance Summary
More informationVirtual Security Server
Data Sheet VSS Virtual Security Server Security clients anytime, anywhere, any device CENTRALIZED CLIENT MANAGEMENT UP TO 50% LESS BANDWIDTH UP TO 80 VIDEO STREAMS MOBILE ACCESS INTEGRATED SECURITY SYSTEMS
More informationHow Data Volume Affects Spark Based Data Analytics on a Scale-up Server
How Data Volume Affects Spark Based Data Analytics on a Scale-up Server Ahsan Javed Awan EMJD-DC (KTH-UPC) (https://www.kth.se/profile/ajawan/) Mats Brorsson(KTH), Vladimir Vlassov(KTH) and Eduard Ayguade(UPC
More informationTuning Cognos ReportNet for a High Performance Environment
Proven Practice Tuning Cognos ReportNet for a High Performance Environment Product(s): Cognos ReportNet Area of Interest: Performance Tuning Cognos ReportNet for a High Performance Environment 2 Copyright
More informationHOW WELL DO YOU KNOW YOUR IT NETWORK? BRIEFING DOCUMENT
HOW WELL DO YOU KNOW YOUR IT NETWORK? BRIEFING DOCUMENT ARE YOU REALLY READY TO EXECUTE A GLOBAL IOT STRATEGY? Increased demand driven by long-term trends of the Internet of Things, WLAN, connected LED
More informationEnabling Efficient and Scalable Zero-Trust Security
WHITE PAPER Enabling Efficient and Scalable Zero-Trust Security FOR CLOUD DATA CENTERS WITH AGILIO SMARTNICS THE NEED FOR ZERO-TRUST SECURITY The rapid evolution of cloud-based data centers to support
More informationQLogic 2500 Series FC HBAs Accelerate Application Performance
QLogic 2500 Series FC HBAs Accelerate QLogic 8Gb Fibre Channel Adapters from Cavium: Planning for Future Requirements 8Gb Performance Meets the Needs of Next-generation Data Centers EXECUTIVE SUMMARY It
More informationIndex. ADEPT (tool for modelling proposed systerns),
Index A, see Arrivals Abstraction in modelling, 20-22, 217 Accumulated time in system ( w), 42 Accuracy of models, 14, 16, see also Separable models, robustness Active customer (memory constrained system),
More informationMassively Parallel Processing. Big Data Really Fast. A Proven In-Memory Analytical Processing Platform for Big Data
Big Data Really Fast A Proven In-Memory Analytical Processing Platform for Big Data 2 Executive Summary / Overview: Big Data can be a big headache for organizations that have outgrown the practicality
More informationSmall verse Large. The Performance Tester Paradox. Copyright 1202Performance
Small verse Large The Performance Tester Paradox The Paradox Why do people want performance testing? To stop performance problems in production How do we ensure this? Performance test with Realistic workload
More informationEfficient Data Center Virtualization Requires All-flash Storage
White Paper Efficient Data Center Virtualization Requires All-flash Storage By Scott Sinclair, Storage Analyst November 2015 This ESG White Paper was commissioned by Pure Storage and is distributed under
More information06-Dec-17. Credits:4. Notes by Pritee Parwekar,ANITS 06-Dec-17 1
Credits:4 1 Understand the Distributed Systems and the challenges involved in Design of the Distributed Systems. Understand how communication is created and synchronized in Distributed systems Design and
More informationReduce Costs & Increase Oracle Database OLTP Workload Service Levels:
Reduce Costs & Increase Oracle Database OLTP Workload Service Levels: PowerEdge 2950 Consolidation to PowerEdge 11th Generation A Dell Technical White Paper Dell Database Solutions Engineering Balamurugan
More informationLecture Topics. Announcements. Today: Advanced Scheduling (Stallings, chapter ) Next: Deadlock (Stallings, chapter
Lecture Topics Today: Advanced Scheduling (Stallings, chapter 10.1-10.4) Next: Deadlock (Stallings, chapter 6.1-6.6) 1 Announcements Exam #2 returned today Self-Study Exercise #10 Project #8 (due 11/16)
More informationSolid Access Technologies, LLC
Newburyport, MA, USA USSD 200 USSD 200 The I/O Bandwidth Company Solid Access Technologies, LLC Solid Access Technologies, LLC Why Are We Here? The Storage Perfect Storm Traditional I/O Bottleneck Reduction
More informationB2W Software Resource Requirements & Recommendations
B2W Software Resource Requirements & Recommendations v2019.2, December 21, 2018 Introduction... 2 Resource Determination Factors... 2 Environment Configuration Options... 3 Considerations... 3 Estimate...
More informationPerformance and Load Testing R12 With Oracle Applications Test Suite
Performance and Load Testing R12 With Oracle Applications Test Suite Deep Ram Technical Director Oracle Corporation Daniel Gonzalez Practice Manager Oracle Corporation Safe Harbor
More informationStellar performance for a virtualized world
IBM Systems and Technology IBM System Storage Stellar performance for a virtualized world IBM storage systems leverage VMware technology 2 Stellar performance for a virtualized world Highlights Leverages
More informationPractical Database Design Methodology and Use of UML Diagrams Design & Analysis of Database Systems
Practical Database Design Methodology and Use of UML Diagrams 406.426 Design & Analysis of Database Systems Jonghun Park jonghun@snu.ac.kr Dept. of Industrial Engineering Seoul National University chapter
More informationChanging the way companies run their data centers
Infrastructure Management & Monitoring for Business-Critical Continuity TM Changing the way companies run their data centers The Aperture TM Suite Optimize performance of your data center without COmpromising
More informationIntroduction. Application Versions. Virtual Machine Defined. Other Definitions. Tech Note 656 Building Wonderware Solution Architectures on VMware
Tech Note 656 Building Wonderware Solution Architectures on VMware All Tech Notes, Tech Alerts and KBCD documents and software are provided "as is" without warranty of any kind. See the Terms of Use for
More informationMULTIPROCESSORS AND THREAD-LEVEL. B649 Parallel Architectures and Programming
MULTIPROCESSORS AND THREAD-LEVEL PARALLELISM B649 Parallel Architectures and Programming Motivation behind Multiprocessors Limitations of ILP (as already discussed) Growing interest in servers and server-performance
More informationThree Key Challenges Facing ISPs and Their Enterprise Clients
Three Key Challenges Facing ISPs and Their Enterprise Clients GRC, enterprise services, and ever-evolving hybrid infrastructures are all dynamic and significant challenges to the ISP s enterprise clients.
More informationContents Overview of the Performance and Sizing Guide... 5 Architecture Overview... 7 Performance and Scalability Considerations...
Unifier Performance and Sizing Guide for On-Premises Version 17 July 2017 Contents Overview of the Performance and Sizing Guide... 5 Architecture Overview... 7 Performance and Scalability Considerations...
More informationFour-Socket Server Consolidation Using SQL Server 2008
Four-Socket Server Consolidation Using SQL Server 28 A Dell Technical White Paper Authors Raghunatha M Leena Basanthi K Executive Summary Businesses of all sizes often face challenges with legacy hardware
More informationMULTIPROCESSORS AND THREAD-LEVEL PARALLELISM. B649 Parallel Architectures and Programming
MULTIPROCESSORS AND THREAD-LEVEL PARALLELISM B649 Parallel Architectures and Programming Motivation behind Multiprocessors Limitations of ILP (as already discussed) Growing interest in servers and server-performance
More informationModeling and Simulation (An Introduction)
Modeling and Simulation (An Introduction) 1 The Nature of Simulation Conceptions Application areas Impediments 2 Conceptions Simulation course is about techniques for using computers to imitate or simulate
More informationBusiness Benefits of Policy Based Data De-Duplication Data Footprint Reduction with Quality of Service (QoS) for Data Protection
Data Footprint Reduction with Quality of Service (QoS) for Data Protection By Greg Schulz Founder and Senior Analyst, the StorageIO Group Author The Green and Virtual Data Center (Auerbach) October 28th,
More informationWhite Paper. Low Cost High Availability Clustering for the Enterprise. Jointly published by Winchester Systems Inc. and Red Hat Inc.
White Paper Low Cost High Availability Clustering for the Enterprise Jointly published by Winchester Systems Inc. and Red Hat Inc. Linux Clustering Moves Into the Enterprise Mention clustering and Linux
More informationLecture 1: January 22
CMPSCI 677 Distributed and Operating Systems Spring 2018 Lecture 1: January 22 Lecturer: Prashant Shenoy Scribe: Bin Wang 1.1 Introduction to the course The lecture started by outlining the administrative
More informationComposable Infrastructure for Public Cloud Service Providers
Composable Infrastructure for Public Cloud Service Providers Composable Infrastructure Delivers a Cost Effective, High Performance Platform for Big Data in the Cloud How can a public cloud provider offer
More informationNetworking Strategy and Optimization Services (NSOS) 2010 IBM Corporation
Networking Strategy and Optimization Services (NSOS) Agenda Network Strategy and Optimization Services (NSOS) Overview IBM NSOS NAO Offerings Model IBM NSOS NIO Offerings Model Why IBM Lot of specialist
More informationSoftNAS Cloud Performance Evaluation on Microsoft Azure
SoftNAS Cloud Performance Evaluation on Microsoft Azure November 30, 2016 Contents SoftNAS Cloud Overview... 3 Introduction... 3 Executive Summary... 4 Key Findings for Azure:... 5 Test Methodology...
More informationTreinamento em Análise Quantitativa & Planejamento de Capacidade. Virgilio A. F. Almeida
Treinamento em Análise Quantitativa & Planejamento de Capacidade Virgilio A. F. Almeida DATAPREV Rio de Janeiro 27 Novembro de 2009 Módulo #3a Departamento de Ciência da Computação Universidade Federal
More informationFAQ: Database Development and Management
Question 1: Are normalization rules followed exclusively in the real world? Answer 1: Unfortunately, the answer to this question is no. Database design and development do not have hard and fast rules,
More informationPeer Software and Scality - A Distributed File System Approach to Scale-out Storage
Peer Software and Scality - A Distributed File System Approach to Scale-out Storage Contents Introduction - What We All Want........................... 2 Why Cloud Gateways................................
More informationPerformance, 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 informationLecture 24: Virtual Memory, Multiprocessors
Lecture 24: Virtual Memory, Multiprocessors Today s topics: Virtual memory Multiprocessors, cache coherence 1 Virtual Memory Processes deal with virtual memory they have the illusion that a very large
More informationTuning WebHound 4.0 and SAS 8.2 for Enterprise Windows Systems James R. Lebak, Unisys Corporation, Malvern, PA
Paper 272-27 Tuning WebHound 4.0 and SAS 8.2 for Enterprise Windows Systems James R. Lebak, Unisys Corporation, Malvern, PA ABSTRACT Windows is SAS largest and fastest growing platform. Windows 2000 Advanced
More informationBECOME A LOAD TESTING ROCK STAR
3 EASY STEPS TO BECOME A LOAD TESTING ROCK STAR Replicate real life conditions to improve application quality Telerik An Introduction Software load testing is generally understood to consist of exercising
More informationCSCI 4717 Computer Architecture
CSCI 4717/5717 Computer Architecture Topic: Symmetric Multiprocessors & Clusters Reading: Stallings, Sections 18.1 through 18.4 Classifications of Parallel Processing M. Flynn classified types of parallel
More informationAn Oracle White Paper March Consolidation Using the Oracle SPARC M5-32 High End Server
An Oracle White Paper March 2013 Consolidation Using the Oracle SPARC M5-32 High End Server Executive Overview... 1 Why Server and Application Consolidation?... 2 Requirements for Consolidation... 3 Consolidation
More informationManaging Data Resources
Chapter 7 Managing Data Resources 7.1 2006 by Prentice Hall OBJECTIVES Describe basic file organization concepts and the problems of managing data resources in a traditional file environment Describe how
More informationAn Introduction to GPFS
IBM High Performance Computing July 2006 An Introduction to GPFS gpfsintro072506.doc Page 2 Contents Overview 2 What is GPFS? 3 The file system 3 Application interfaces 4 Performance and scalability 4
More informationOPTIMIZATION MAXIMIZING TELECOM AND NETWORK. The current state of enterprise optimization, best practices and considerations for improvement
MAXIMIZING TELECOM AND NETWORK OPTIMIZATION The current state of enterprise optimization, best practices and considerations for improvement AOTMP.com The Next Evolution of Telecom Management OVERVIEW As
More informationAnalytical Modeling of Parallel Programs
2014 IJEDR Volume 2, Issue 1 ISSN: 2321-9939 Analytical Modeling of Parallel Programs Hardik K. Molia Master of Computer Engineering, Department of Computer Engineering Atmiya Institute of Technology &
More informationMultiprocessor Support
CSC 256/456: Operating Systems Multiprocessor Support John Criswell University of Rochester 1 Outline Multiprocessor hardware Types of multi-processor workloads Operating system issues Where to run the
More informationDatabase Systems: Design, Implementation, and Management Tenth Edition. Chapter 1 Database Systems
Database Systems: Design, Implementation, and Management Tenth Edition Chapter 1 Database Systems Objectives In this chapter, you will learn: The difference between data and information What a database
More informationMultiple Processor Systems. Lecture 15 Multiple Processor Systems. Multiprocessor Hardware (1) Multiprocessors. Multiprocessor Hardware (2)
Lecture 15 Multiple Processor Systems Multiple Processor Systems Multiprocessors Multicomputers Continuous need for faster computers shared memory model message passing multiprocessor wide area distributed
More informationUnderstanding Managed Services
Understanding Managed Services The buzzword relating to IT Support is Managed Services, and every day more and more businesses are jumping on the bandwagon. But what does managed services actually mean
More informationDon t Plan Capacity When You Should Predict Applications
Don t Plan Capacity When You Should Predict Applications Tim R. Norton Doctoral Candidate Colorado Technical University CMG97 Session 443, December 10, 1997 As application designs incorporate client/server
More informationExtrapolation Tool for Load Testing Results
Extrapolation Tool for Load Testing Results Subhasri Duttagupta, Rajesh Mansharamani Performance Engineering Lab Tata Consulting Services Mumbai, India subhasri.duttagupta@tcs.com, rajesh.mansharamani@tcs.com
More informationChoosing a Processor: Benchmarks and Beyond (S043)
Insight, Analysis, and Advice on Signal Processing Technology Choosing a Processor: Benchmarks and Beyond (S043) Jeff Bier Berkeley Design Technology, Inc. Berkeley, California USA +1 (510) 665-1600 info@bdti.com
More informationBest Practices for Setting BIOS Parameters for Performance
White Paper Best Practices for Setting BIOS Parameters for Performance Cisco UCS E5-based M3 Servers May 2013 2014 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public. Page
More informationOracle Application Server Forms Services 10g (9.0.4) Capacity Planning Guide. An Oracle White Paper November 2004
Oracle Application Server Forms Services 10g (9.0.4) Capacity Planning Guide An Oracle White Paper November 2004 Oracle Application Server Forms Services 10g (9.0.4) Capacity Planning Guide What is in
More informationGPU ACCELERATED DATABASE MANAGEMENT SYSTEMS
CIS 601 - Graduate Seminar Presentation 1 GPU ACCELERATED DATABASE MANAGEMENT SYSTEMS PRESENTED BY HARINATH AMASA CSU ID: 2697292 What we will talk about.. Current problems GPU What are GPU Databases GPU
More informationCapacity Planning for Application Design
WHITE PAPER Capacity Planning for Application Design By Mifan Careem Director - Solutions Architecture, WSO2 1. Introduction The ability to determine or forecast the capacity of a system or set of components,
More informationMultiprocessing and Scalability. A.R. Hurson Computer Science and Engineering The Pennsylvania State University
A.R. Hurson Computer Science and Engineering The Pennsylvania State University 1 Large-scale multiprocessor systems have long held the promise of substantially higher performance than traditional uniprocessor
More informationMultiprocessors & Thread Level Parallelism
Multiprocessors & Thread Level Parallelism COE 403 Computer Architecture Prof. Muhamed Mudawar Computer Engineering Department King Fahd University of Petroleum and Minerals Presentation Outline Introduction
More informationRED HAT ENTERPRISE LINUX. STANDARDIZE & SAVE.
RED HAT ENTERPRISE LINUX. STANDARDIZE & SAVE. Is putting Contact us INTRODUCTION You know the headaches of managing an infrastructure that is stretched to its limit. Too little staff. Too many users. Not
More informationOPTIMIZATION, OPTIMAL DESIGN AND DE NOVO PROGRAMMING: DISCUSSION NOTES
OPTIMIZATION, OPTIMAL DESIGN AND DE NOVO PROGRAMMING: DISCUSSION NOTES MILAN ZELENY Introduction Fordham University, New York, USA mzeleny@fordham.edu Many older texts, with titles like Globally Optimal
More informationDELL EMC DATA DOMAIN SISL SCALING ARCHITECTURE
WHITEPAPER DELL EMC DATA DOMAIN SISL SCALING ARCHITECTURE A Detailed Review ABSTRACT While tape has been the dominant storage medium for data protection for decades because of its low cost, it is steadily
More informationAn Oracle White Paper September Oracle Utilities Meter Data Management Demonstrates Extreme Performance on Oracle Exadata/Exalogic
An Oracle White Paper September 2011 Oracle Utilities Meter Data Management 2.0.1 Demonstrates Extreme Performance on Oracle Exadata/Exalogic Introduction New utilities technologies are bringing with them
More informationDistributed File Systems Issues. NFS (Network File System) AFS: Namespace. The Andrew File System (AFS) Operating Systems 11/19/2012 CSC 256/456 1
Distributed File Systems Issues NFS (Network File System) Naming and transparency (location transparency versus location independence) Host:local-name Attach remote directories (mount) Single global name
More informationReduce Latency and Increase Application Performance Up to 44x with Adaptec maxcache 3.0 SSD Read and Write Caching Solutions
Reduce Latency and Increase Application Performance Up to 44x with Adaptec maxcache 3. SSD Read and Write Caching Solutions Executive Summary Today s data centers and cloud computing environments require
More informationW H I T E P A P E R U n l o c k i n g t h e P o w e r o f F l a s h w i t h t h e M C x - E n a b l e d N e x t - G e n e r a t i o n V N X
Global Headquarters: 5 Speen Street Framingham, MA 01701 USA P.508.872.8200 F.508.935.4015 www.idc.com W H I T E P A P E R U n l o c k i n g t h e P o w e r o f F l a s h w i t h t h e M C x - E n a b
More information#DeloitteInnovation: In-Time How efficiently do you use your SAP HANA?
#DeloitteInnovation: In-Time How efficiently do you use your SAP HANA? Deloitte In-Time in a Nutshell In-Time is the first and only SAP HANA optimization software that can analyze the effectiveness of
More informationLecture 9: MIMD Architecture
Lecture 9: MIMD Architecture Introduction and classification Symmetric multiprocessors NUMA architecture Cluster machines Zebo Peng, IDA, LiTH 1 Introduction MIMD: a set of general purpose processors is
More informationTerms, Methodology, Preparation, Obstacles, and Pitfalls. Vulnerability Assessment Course
Terms, Methodology, Preparation, Obstacles, and Pitfalls Vulnerability Assessment Course All materials are licensed under a Creative Commons Share Alike license. http://creativecommons.org/licenses/by-sa/3.0/
More informationTHREAT REPORT Medical Devices
THREAT REPORT Medical Devices Detailed analysis of connected medical devices across 50 hospitals in 2017 THREAT REPORT In this Threat Report Introduction 3 About This Report 3 Device Deployments 4 Most
More informationDistributed Systems LEEC (2006/07 2º Sem.)
Distributed Systems LEEC (2006/07 2º Sem.) Introduction João Paulo Carvalho Universidade Técnica de Lisboa / Instituto Superior Técnico Outline Definition of a Distributed System Goals Connecting Users
More informationThe Benefits of Wireless Infrastructure Management in the Cloud
WHITE PAPER The Benefits of Wireless Infrastructure Management in the Cloud High Performance Wireless Networks The Benefits of Wireless Infrastructure Management in the Cloud How the cloud maximizes IT
More informationDisco. CS380L: Mike Dahlin. September 13, This week: Disco and Exokernel. One lesson: If at first you don t succeed, try try again.
Disco CS380L: Mike Dahlin September 13, 2007 Disco: A bad idea from the 70 s, and it s back! Mendel Rosenblum (tongue in cheek) 1 Preliminaries 1.1 Review 1.2 Outline 1.3 Preview This week: Disco and Exokernel.
More informationQLIKVIEW SCALABILITY BENCHMARK WHITE PAPER
QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER Measuring Business Intelligence Throughput on a Single Server QlikView Scalability Center Technical White Paper December 2012 qlikview.com QLIKVIEW THROUGHPUT
More informationMassive Scalability With InterSystems IRIS Data Platform
Massive Scalability With InterSystems IRIS Data Platform Introduction Faced with the enormous and ever-growing amounts of data being generated in the world today, software architects need to pay special
More informationCloud Computing: Making the Right Choice for Your Organization
Cloud Computing: Making the Right Choice for Your Organization A decade ago, cloud computing was on the leading edge. Now, 95 percent of businesses use cloud technology, and Gartner says that by 2020,
More information