Developing Faster Applications – An Overview - 60 Minutes
This module starts with a discussion of why optimization is important, and reviews a successful methodology used by Intel engineers when working with large commercial applications. You will learn about the questions you need to ask before you go down the path of optimizing the application, and find out direct ways to get these questions answered. Specific focus will be on CPU utilization and I/O dependencies, and a review of a case study that shows how the necessary data is gathered and put to successful use. Another case study will show the methodology used in gradually applying new optimization methods to a benchmark which results in cumulative speedups on different problem sizes.
Topics include:- Benefits of Optimization
- Optimization Methodologies
- Tools which can improve application performance
Speaker:Charles Congdon, Intel Senior Software Engineer
Software Tools & Technologies for Multi-Core Architecture for Architects and Developers - 60 Minutes
This Net Seminar will begin with a short history of the architectures of microprocessors, highlighting how more parallelism has been added in each generation, ending with today's dual and multi-core processors. As the processor architecture evolves, so must the software architecture of applications. We will highlight the opportunities, and challenges, in preparing for a ubiquitous multi-core world.
Topics include:- What is multi-threading?
- How do I thread my application?
- What tools can assist me in developing a multi-threaded application?
- Will I need that much compute power
Speaker:Kevin J. Smith, Director, Intel Compiler Lab
Multithreaded Programming Quickstart - 60 Minutes
This Net Seminar explains the benefits of adding parallelism to your application. You will get a high-level understanding of the methods for locating areas within an application to add parallelism. The presentation also highlights some of the common issues which can cause incorrect behavior or poor performance in a threaded application. It concludes with suggestions for ways to incrementally add concurrency without having to completely re-implement your application.
Topics include:- High level methodologies for adding parallelism to an application
- Typical problems found in threaded code
- Tools for debugging threaded applications
Speaker:Charles Congdon, Intel Senior Software Engineer
The Process of Threading: Multithreading a Sequential Application for a Multi-core World – 60 Minutes
This training will step the attendee through the process of analyzing and then threading a sequential application for performance. We will begin with a discussion of threads and threading models. After our threading introduction, we will discuss analysis tools that help a developer identify where the application can benefit from threading and how to estimate the potential gains. Next, we will move to a discussion of debugging and tuning threads for performance. This webcast will feature demonstrations of how to use Intel® Thread Checker for identifying race conditions and Intel® Thread Profiler for finding thread performance bottlenecks. We relate performance back to Amdahl’s law for estimations and verification of work accomplished.
Speaker:Dr. David Mackay, Manager, Consulting and Technical Support Group for Intel® Threading Tools.
Tuning Threads for Performance: Improving Software Performance with the Intel® Thread Profiler
This presentation drills down on the topic of threading for performance by utilizing Intel® Thread Profiler.
Topics include:- Tuning threaded software.
- Correcting workload imbalances across threads.
- Identifying software objects and contention points.
- Minimizing locks and barriers
- Using the timeline view to develop & design better software.
Speaker:Douglas Armstrong, Manager Threading Tools Development
Intel® Threading Building Blocks: Scalable Programming for Multi-core – 60 Minutes
This presentation will introduce the simplest features of the new parallel programming model – setting up generic parallel algorithms.
Topics include:- Issues for getting real multi-core speedup
- Basic parallel patterns: forall, reduction, while and pipeline
- The power of generic programming for parallelism
- Incorporating the new programming model into your software
Speaker:Arch Robinson, Principal Engineer, Intel Corporation
Multithreading for Multi-Core: A Focus on Managed Runtime Environments – 60 Minutes
This Net Seminar will introduce the basics of multi-core platforms as well as the core concepts of multi-threading in MRTEs. We will discuss techniques and tools used to build multi-threaded MRTE applications with ease and will work through a couple of real-life scenarios and applications to identify common threading problems and ultimately to maximize application performance.
After viewing this Net Seminar you will:
- Understand why multi-threading is essential for both performance and functionality on today’s multi-core platforms
- Appreciate how managed runtime environments simplify your efforts in implementing multi-threaded applications
- Understand different managed threading techniques and their APIs
- Know how to overcome the difficulties of creating high-performance multi-threaded applications with the aid of Intel’s Software Development Products
Speaker:Rami Radi, Sr. Software Application Engineer, Intel Corporation
Programming Multithreaded Dynamic Work Assignment Models – 60 Minutes
After attending this Net Seminar, participants will understand implementation of dynamic task allocation strategies with explicit threading models and the importance of proper planning and examination of all possible thread interactions.
Speaker:Clay Breshears, Sr. Course Architect, Intel Software College
Making Persistence Work: Using Thread Local Storage – 60 Minutes
After attending this Net Seminar, participants will understand the functions used to declare, use, and remove Thread Local Storage in both Windows threads and Pthreads also how to effectively make use of Thread Local Storage in multithreaded applications and thread-safe libraries.
Speaker:Clay Breshears, Sr. Course Architect, Intel Software College
Porting Applications to 64-bit - 60 Minutes
This Net Seminar begins with a discussion of when (and if) it is desirable to port your application to take advantage of 64-bit address spaces. It details Intel's 64-bit hardware and software support and focuses on some of the problems and pitfalls developers may encounter along the way. It presents a detailed methodology to help minimize the pain felt during such a transition as well as a discussion of the changes in the APIs and tools in 64-bit Windows* and Linux*. The presentation concludes by touching on some of the performance problems encountered in a 64-bit application, and the tools used to ameliorate them.
Topics include:- Why move to 64-bits?
- Intel 64-bit hardware and software support
- 64-bit data models
- Minimizing the porting effort
- API changes
- Issues effecting 64-bit code size and performance
Speaker:Charles Congdon, Intel Senior Software Engineer
Making Mashups Mobile – Presented by Intel, Amazon.com and ElephantDrive - 90 minutes
Intel Corporation, Amazon.com and ElephantDrive have cooperated to implement a mashup that improves the mobile aspects of existing web solutions. Jeff Barr, an Amazon Web Services Evangelist, will provide an overview of Amazon Web Services and show you the possibilities created by these innovative offerings. Lester Memmott, a software architect in Intel’s Software and Solutions Group, will give an overview of the Intel® Mobile Platform SDK and its benefits. Mike Fisher, Co-founder of ElephantDrive will be presenting a real-life example of how their mashup, with the Intel’s Mobile SDK and Amazon S3 APIs have improved their customers’ user experience for online back-up and storage. We will also take a look into the code ElephantDrive used in their implementation and discuss the techniques and technologies available to develop more robust mashups that deliver business value.
Learn more about:- Web technologies and techniques for mashups
- Mashup/Web 2.0 apps for mobile users
- Writing business quality mashups
Speakers:Lester Memmott, Software Architect, Software & Solutions Group, Intel
Michael Fisher, Co-founder, ElephantDrive
Jeff Barr, Evangelist, Amazon Web Services
How to Build Better Remote Manageability with Intel® Active Management Technology - 60 Minutes
A major barrier to greater IT efficiency & manageability has been removed by Intel® Active Management Technology (Intel® AMT). This technology gives developers an out of band channel to the system that is not related to the host software. Using built-in platform capabilities, developers can now build better remote manageability and security into your applications.
This training discusses developing using Intel® AMT in system management applications. It presents the business reasons for adoption and the relevant use cases using this technology to realize the value, describes the architecture of Intel® AMT and delineates the resources available for developers and software vendors, including the software development kit (SDK) and reference design (RDK).
After this NetSeminar, you will be able to:- Appreciate how Intel® AMT extends software-only manageability approaches with out-of-band capabilities
- Understand the hardware and software architecture of Intel® AMT
- Know how the SDK and RDK can help build applications that take advantage of Intel® AMT
Speaker:Bryan Duerk, Intel Sr. Developer Evangelist
Bryan Duerk is the Senior Developer Evangelist for Enterprise and Mobile Clients in Intel’s Global Developer Relations Group. In his current role, Bryan is responsible for the Intel® Active Management Technology Scale Program. He joined Intel in 1997 and has worked previously at Intel as a Software Architect in Intel® Solution Services consulting to Fortune 500 companies.
Design Patterns for Parallel Programming: Learning to Think Parallel - 60 Minutes
A specialized community of software developers has been writing parallel programs for over 25 years and has compiled a considerable body of knowledge about parallel programming. Capturing this knowledge and transferring it to the programming community at large, however, has proven difficult. To address this problem we created a Design Pattern Language for parallel programming (Patterns for Parallel Programming, Mattson, Sanders and Massingil, Addison Wesley, 2004). With this pattern language, we can help transform experienced software engineers into parallel software engineers.
In this NetSeminar, we will provide an overview of this design pattern language and provide a few cases studies of how it is used in practice.
Speaker:Dr. Tim Mattson, Principal Engineer, Intel's Application Research Laboratory
Dr. Tim Mattson is a Principle Engineer at Intel's Application Research Laboratory. He joined Intel in 1993. Among his many roles at Intel, he was applications manager for the ASCI teraFLOPS project, helped create OpenMP, founded the Open Cluster Group (OSCAR), and launched Intel's programs in computing for the Life Sciences. Dr. Mattson earned a PhD. for his work on quantum molecular scattering theory (UCSC, 1985). This was followed by a Post-doc at Caltech where he worked on the Caltech/JPL hypercubes. Currently, Dr. Mattson is conducting research on performance modeling for future multi-core microprocessors and how different programming models map onto these systems.
Open MP: At the Frontier - 60 Minutes
OpenMP continues to evolve toward a general-purpose threading model, and for many will become the model of choice for multi-core platforms. Past emphasis has been on data parallel loops, but there are still many opportunities for concurrent processing and task parallel computation.
This NetSeminar will address new, key components of the latest 3.0 specification draft -- in particular, we will focus on the task queuing extension.
Those attending this NetSeminar will learn:- What's new in OpenMP 3.0
- Parallelizing beyond loops: work queuing for task parallel jobs and list processing
Speaker:Dr. Tim Mattson, Principal Engineer, Intel's Application Research Laboratory
Dr. Tim Mattson is a Principle Engineer at Intel's Application Research Laboratory. He joined Intel in 1993. Among his many roles at Intel, he was applications manager for the ASCI teraFLOPS project, helped create OpenMP, founded the Open Cluster Group (OSCAR), and launched Intel's programs in computing for the Life Sciences. Dr. Mattson earned a PhD. for his work on quantum molecular scattering theory (UCSC, 1985). This was followed by a Post-doc at Caltech where he worked on the Caltech/JPL hypercubes. Currently, Dr. Mattson is conducting research on performance modeling for future multi-core microprocessors and how different programming models map onto these systems.
The Service Oriented Enterprise (SOE), An Architect’s View - 90 minutes
This Net Seminar introduce some of the emergent technologies and architecture such as SOA, Grid, and Virtualization under the umbrella of Service Oriented Enterprise (SOE). Topics covered include the Service Oriented Enterprise Framework and its components: SOA, SOM, SOI. Learn how technology innovations developed by Intel and its industry partners can provide a model for deployment of the SOE Enterprise.
Learn more about:- SOA best practices
- Real world examples drawn from Intel Case studies from the energy, banking and finance, and healthcare industries
Speaker:Mark Chang, Principal Strategist, Intel Customer Solution Group
Architecting a UNIX* to Linux* Solution Strategy – Co-Presented with IBM - 90 minutes
Targeted towards enterprise architects, this session provides a better understanding of successful Linux* solution migration methodologies developed by Intel and IBM. This Net Seminar will enable architects to appreciate the technical challenges, enhance their knowledge of alternative solutions available on open platforms, and enable them to minimize TCO and enhance ROI as they migrate enterprise solutions.
Learn more about:- Enterprise solution methodologies for Unix* to Linux migration
- How to minimize TCO and enhance ROI
- Using open source platforms
- Migration tools and services available from Intel and IBM
Speakers:Nitin Gupta, Intel Solution Services Senior Consulting Engineer
Michael Prentice, IBM Senior Technical Solution Manager
Moving to Enterprise Utility Computing: Next-Generation Enterprise IT Infrastructure Available Today - 90 Minutes
Intel Corporation and Platform Computing present a Net Seminar on Enterprise Utility Computing (EUC), a new approach to designing robust and flexible IT infrastructure based on technologies available now.
In today's environment, network, computing, and storage resources need to be dynamically managed and aligned to support business services. The need to share resources goes to the heart of computing infrastructure as a major issue for corporate datacenters. EUC addresses not only the management of physical and virtual computing resources by enforcing organizational and business policies, but also provides a framework for providing IT infrastructure as services with metering, charge back, and reporting capabilities.
This vision of a next-generation IT environment that copes with ever-changing business requirements by applying available technologies such as grid, virtualization, and new application environments based on Services Oriented Architecture (SOA).
Speakers:John Benchoff, Product Manager, Platform Computing
Parviz Peiravi, Lead Architect, Financial Services Industry, Customer Solutions Group, Solutions Architecture and Design, Intel Corporation
Grid Computing for Architects: Unlocking the Value of Grid with IBM and Intel-based Server Architecture – Co-presented with IBM - 90 Minutes
Organizations across the world are realizing business value from Grid computing. Find out how and why the emergence of Grid computing as an enterprise, distributed architecture enables companies to better align resources to business objectives, accelerates their time to market to gain competitive advantage, and provides an infrastructure foundation for innovation and collaboration across the enterprise. Hear from both Intel and IBM's leading grid experts on how to get started with Grid as well as key Grid technologies and their applicability to your enterprise. See how Grid should be a key part of your enterprise virtualization strategy. And understand the alignment of Grid as an infrastructure technology with service oriented architectures and the importance of picking the right platforms, technologies and partners to realize business success with Grid. Further, learn about the tools, solutions and best practices to speed your development, improve performance, and appropriately integrate security.
Speakers:Robert Fogel, Intel Director of World-Wide Grid Strategy and Business Development; vice-chairman of the Global Grid Forum (GGF).
Matt Haynos, Program Director, Grid Technology and Strategy, IBM
Building a Solid Grid and Virtualization Foundation with Intel and IBM
In this Net Seminar, we will highlight how data centers are evolving to Virtual Enterprises and how business environment and technological advances are helping fuel the adoption of Virtualization in all layers of the enterprise.
Speakers:Scott Klasing, IBM, Lead Architect for Americas Grid Services
Hans–Christian Hoppe, Intel, Director, Grid Technology Lab
Enterprise Benchmarking Primer: An Overview for Enterprise Architects - 60 Minutes
Designed for decision makers, this course demystifies "enterprise benchmarking".
Topics covered include:- Characteristics of a good benchmark
- Common enterprise benchmarks
- How to select a relevant benchmark
- How to use a benchmark for decision making while avoiding the common pitfalls
- How to benchmark your new “virtualized infrastructure environment” that uses virtual machines and virtualization technologies
- Intel software tools available to help you with your enterprise benchmarking initiatives
- A review of available industry vertical specific benchmarks
Speaker:Nitin Gupta, Intel Solution Services Senior Consulting Engineer
