Oracle Grid Computing An Oracle Business White Paper February 2005 Oracle Grid Computing EXECUTIVE OVERVIEW The speed and unpredictability of business cycles have pushed many enterprises to the limit of manageability. Organizations need to become more adaptable but often their information systems are slow to respond. At the same time, these enterprises want to gain greater efficiency from their IT systems and lower the cost of computing. Grid computing is a new IT architecture that adapts to your changing business needs. Even better, grid computing revolutionizes IT economics. With enterprise grid computing you can build a powerful data center with a variable cost structure that can be adjusted as needed. Grid computing is rapidly growing in popularity. Forrester Research reports that 37% of enterprises are piloting, rolling out or have implemented some form of grid computing.1 IDC calls grid computing the fifth generation of computing, after client-server and multi-tier.2 Oracle Corporation has introduced Oracle 10g infrastructure software, the first specifically designed for enterprise grid computing. Oracle 10g software also protects your technology investments today as your data center evolves to enterprise grid computing. Leading businesses, such as Dell and the Chicago Stock Exchange, have begun deploying enterprise grids using Oracle.3 Oracle Grid Computing Page 2 INTRODUCTION Volatility is now a permanent feature of business life. While change is normal, the growing speed and unpredictability of change—which is increasingly technologydriven—have pushed many enterprises to the limit of manageability. Businesses have to develop more adaptive techniques to create value, no matter how unstable the economic environment may be. Enterprise grid computing is about using technology as a way of delivering business value quickly and reliably, despite changes in the environment. Oracle 10g infrastructure software is the first specifically designed for the next generation of computing—enterprise grids. Oracle 10g infrastructure software adapts to your changing business needs. That means you can spend more time thinking about how to run your business, knowing that Oracle 10g software will respond with the reliable, secure performance that your applications need, so you get the answers you need. No other infrastructure software can do this. Even better, Oracle grid computing revolutionizes IT economics. Faced with tight budgets, every business needs to cut IT costs, improve staff productivity, and reduce costly downtime. With Oracle grid computing you can build a powerful data center with a variable cost structure that you can adjust as needed. Oracle 10g infrastructure software also protects your technology investments today as your data center evolves to enterprise grid computing. At each step on the way, you can lower costs and improve business performance, ensuring a solid return on investment as you go. WHAT IS GRID COMPUTING? In simplest terms, grid computing is the pooling of all IT resources into a single set of shared services for all enterprise computing needs. Grid computing infrastructure continually analyzes demand for resources and adjusts supply accordingly. You don’t have to worry about where your data resides, or what computer processes your request. You request information or computing power and have it delivered—as much as you want, whenever you want. This is analogous to the way electric utilities work, in that you don’t know where the generator is, or how the electric grid is wired. You just ask for electricity and you get it. Two core tenets uniquely distinguish grid computing from other styles of computing, such as mainframe, client-server, or multi-tier: virtualization and provisioning. • With virtualization, individual resources (e.g. computers, disks, application components and information sources) are pooled together by Oracle Grid Computing Page 3 type then made available to consumers (e.g. people or software programs) through an abstraction. Virtualization means breaking hard-coded connections between providers and consumers of resources, and preparing a resource to serve a particular need without the consumer caring how that is accomplished. • With provisioning, when consumers request resources through a virtualization layer, behind the scenes a specific resource is identified to fulfill the request and then it’s allocated to the consumer. Provisioning as part of grid computing means that the system determines how to meet the specific need of the consumer, while optimizing performance of the system as a whole. The specific ways in which information, application, or infrastructure resources are virtualized and provisioned are specific to the type of resource, but the concepts apply universally. Similarly, the specific benefits derived from grid computing are particular to each type of resource, but all share the characteristics of better quality, lower costs and increased flexibility. BENEFITS OF GRID COMPUTING Grid computing provides the following benefits: Flexibility to meet changing business needs Highest quality of service at the lowest cost Investment protection and rapid return Enterprise grid computing pools IT resources into a single shared service that continually analyzes demand for resources and adjusts supply accordingly. Flexibility to meet changing business needs Grid computing enables an organization to tie its business architecture, through service level agreements, to its IT architecture. Most applications today use resources from just one server in a single location. When business processes change, new server capacity may need to be purchased, new integration software written, and more testing done. The infrastructure may take a very long time to change to reflect new business requirements. Grid computing introduces sophisticated workload management capabilities that make it possible for business processes to share resources. Data processing capacity can be added or removed on demand, and resources within a location can be dynamically provisioned. Web services can quickly integrate applications to create new business processes. As a result, your IT infrastructure can respond immediately to changes in your business needs. Oracle Grid Computing Page 4 High quality of service at a low cost With a grid computing architecture you can quickly and easily create a large-scale computing infrastructure from inexpensive, off-the-shelf components like server blades and commodity storage.. With a grid computing architecture, costs are fundamentally lower. You can quickly and easily create a large-scale computing infrastructure from inexpensive, off-theshelf components like server blades and commodity storage. You can also reuse application resources by combining Web services into different business processes. Grid computing’s virtualization and provisioning deliver the highest quality of service. You get high performance and scalability because all computing resources can be flexibly allocated to applications as needed. Grid computing tolerates all the common sources of system downtime including hardware, software, network and operator error. By centralizing resources and treating them as a single unit, grid computing makes it possible to create a far more secure infrastructure than any other architecture. Investment protection and rapid return Typically, grid solutions from other vendors require custom applications and completely new infrastructure. With Oracle you can move from your existing infrastructure to grid computing at your own pace. At each step along the way you can achieve higher levels of efficiency, lower operating costs, and a more rapid return on investment. And you can run all your existing applications. Oracle grid computing is platform independent. With the Oracle 10g software, you can pool servers and storage into a flexible, on-demand computing resource for all enterprise computing needs. Any kind of system—mainframes, UNIX, Windows, and Linux servers—can be used to build an enterprise grid based on the Oracle 10g infrastructure software. Oracle 10g software can even integrate applications and data held in non-Oracle systems into a grid. THE DIMENSIONS OF GRID COMPUTING The principles of grid computing can be applied to all your IT resources: infrastructure, applications, and information. Infrastructure Infrastructure grid resources include hardware resources such as storage, processors, memory, and networks as well as software designed to manage this hardware, such as databases, storage management, system management, application servers, and operating systems. Oracle Grid Computing Page 5 Virtualize Infrastructure Resources: Pool hardware and systems software into a single virtual resource. Provision Infrastructure Resources: Allocate capacity on demand based on policies to meet individual needs and optimize the system as a whole. Virtualization and provisioning of infrastructure resources mean pooling resources together and allocating to the appropriate consumers based on policies. For example, one policy might be to dedicate enough processing power to a Web server that it can always provide sub-second response time. This rule could be fulfilled in different ways by the provisioning software in order to balance the requests of all consumers. Treating infrastructure resources as a single pool and allocating those resources on demand saves money by eliminating underutilized capacity and redundant capabilities. Managing hardware and software resources holistically reduces the cost of labor and the opportunity for human error. Spreading computing capacity among many different computers and spreading storage capacity across multiple disks and disk groups removes single points of failure so that if any individual component fails, the system as a whole remains available. Grid computing affords the option to use smaller individual hardware components, such as blade servers and low cost storage, which enables incremental scaling and reduces the cost of each individual component, thereby giving companies more flexibility and lower cost. Infrastructure is the dimension of grid computing that is most familiar and easy to understand, but the same concepts apply to applications and information. Applications Application resources in the grid are the business logic and process flows that are the components of application software. These resources may be in packaged applications or custom applications, and they may be written in any programming language. For example, the software that takes an order from a customer and sends an acknowledgement, the process that prints payroll checks, and the logic that routes a particular customer call to a particular agent are all application resources. Virtualization of application resources involves publishing application components as services for use by multiple consumers, which may be people or processes. Orchestration is the provisioning of those services into more powerful business flows. For example, the business flow of preparing for a new employee could be orchestrated by combining the following processes: create a user in the HR system, create a new email account, order a computer, set up phone and voicemail, and deliver a welcome kit. The key is that these individual processes can be exposed as services completely independently from each other then integrated easily. In the same way that grid computing enables better reuse and more flexibility of IT infrastructure resources, grid computing also treats application components as resources. Then, by publishing and orchestrating those resources into more complex business flows, grid computing enables greater reuse of application functionality and more flexibility in changing and building new composite applications. Treating application logic as another resource in the grid enables better software reuse and creates applications that are easier to change. Oracle Grid Computing Page 6 Applications that are orchestrated from published services are capable of viewing activities in a business as a single whole, so that processes are standardized across geography and business units and processes are automated end-to-end. This generates more reliable business processes and lowers cost through increased automation and reduced variability. The characteristics and benefits described for application grid computing are exactly the characteristics and benefits of a Service-Oriented Architecture (SOA). The foundation of SOA is a set of services—independent, well-defined encapsulations of software functionality that can be invoked over a network using heterogeneous platforms and execution environments. SOA connects these independent services toward a larger purpose. Oracle’s recommended way to implement SOA is using XML-based Web services standards. Web services are successful where earlier distributed computing architectures failed for three reasons: standards, broad adoption, and loose coupling. First, the Web services standards that define common interfaces (i.e. WSDL) and a common message construct (i.e. SOAP) are simple. Also, standards for surrounding functionality, such as business flows and industry-specific data representations, are rapidly being defined. Second, Web services standards use underlying internet standards, such as HTTP, that are already pervasive, which reduces the barrier to adoption. Furthermore, Web services standards have been incorporated broadly into packaged software and adopted by companies across many industries, thus reinforcing the value of Web service adoption to others. Finally, XML enables Web services to be more loosely coupled and therefore less brittle in the face of change than alternative distributed architectures, such as distributed objects. XML-based Web services facilitate a document-style exchange of information which can be asynchronous and largely self-describing. XML further accommodates change and reduces brittleness by allowing content that is not understood by the service consumer to be ignored. The cost of the flexibility of XML is that more data is exchanged over the network than would be necessary in a more tightly coupled environment. For this reason, the ideal Web services implementation of SOA exposes relatively coarse-grained services and uses a more tightly coupled programming methodology, such as J2EE or .NET, for implementation and communication within a given service. Many of the vendors interested in grid computing have been focused on grid resource management as it applies to computers, storage, networking, and operating systems. Meanwhile, middleware vendors and applications vendors have been separately pursuing the grid computing ideals in the domain of applications, under the designation of Web services and Service-Oriented Architecture. These two worlds are now converging, however. Web services standards have emerged as the common basis to support language independent message delivery for all types XML-based web services have emerged as the standard of choice for implementing service oriented architecture and, thus, for grid computing. Oracle Grid Computing Page 7 of resources. New grid standards, defined by the Global Grid Forum, are building on Web services standards as defined by W3C and OASIS. Information The third dimension to grid computing, after infrastructure and applications, is information. Information grid resources include all data in the enterprise and all metadata required to make that data meaningful. This data may be structured, semistructured, or unstructured, stored in any location, such as databases, local file systems, or email servers, and created by any application. The way these information resources are used together in a grid is similar to the way infrastructure and applications resources are used together in a grid. The infrastructure grid exploits the power of the network to allow multiple servers or storage devices to be combined toward a single task, then easily reconfigured as needs change. A Service-Oriented Architecture, or an applications grid, enables independently developed application resources to be combined into larger business processes, then adapts as needs change without breaking other parts of the composite application. Similarly, the information grid provides a way for information resources to be connected with related information resources to greater exploit the value of the inherent relationships among information. New connections between information are made as needed. Many of the building blocks required to build an information grid exist today, but true information grids are still largely visionary. The vision for the information grid is shared as the vision of the Semantic Web. As defined by the W3C, “the Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries.”4 The primary missing link to effectively sharing and reusing data on the Web is the lack of machine-readable standards for semantics (i.e. meanings) of HTML-based content. Thus, the Semantic Web is often associated with specific XML-based standards for semantics, such as Resource Description Framework (RDF) and Web Ontology Language (OWL). As with the Semantic Web, the information grid requires metadata describing the meaning of data and relationships among data elements so that information can be reused in new ways. In an enterprise, the best way to make information reusable depends on its source and format, the predictability of its content, and the uses for which the information should be optimized. Many technologies facilitate new relationships between information resources and contribute toward the information grid. Oracle Grid Computing Page 8 What do relational databases, XML, and Google have in common? They all serve as base technologies for grid computing by virtualizing information representation and access. For example, the relational database was one such early metadata technology. Unlike its predecessors, the network database and hierarchical database, in which all relationships between data had to be predetermined, the relational database enabled flexible access to a general-purpose information resource. Similarly, XML serves as a base technology for managing information in a grid because it provides a powerful way to represent information along with metadata, which breaks the hard link between information and a specific application used to create and view information. By crawling the content of the Web, Google’s technology deduces metadata about Web-based information that allows it to be queried and used in new ways. Enterprise search servers provide the same opportunity for reuse of content within a company intranet. On the Web, where content ownership is broadly distributed, finding and exploiting all the inherent relationships between individual information resources requires modeling semantics in metadata, then creating tools to utilize the metadata to find and exploit new connections. Within an enterprise, most of the highest value information can be managed holistically, which greatly improves prospects for exploiting all possible relationships among information. In a simple example, a particular product order may be further exploited for historical analysis in combination with other product orders, other requests from the same type of customer, emails from that specific customer, or other products from the same supplier. The semantics to enable fuller exploitation of such structured data may be embedded in transformation or data cleansing tools or in application suites. These types of proprietary methods for representing metadata can be the best solution for enterprises looking to fully exploit their own proprietary information. When centralization is not an option for reasons of ownership, complexity, or legacy, some form of enterprise information integration will be required to relate information from disparate applications and formats. For some types of enterprise information that are less predictable, centralization and strict structure may be not only impractical but also undesirable. For example, email and instant messages cannot be generated like product orders from structured, form-based inputs which would allow the information to be fully modeled semantically, yet these sources are rich underexploited information assets. More flexible ways to capture and deduce as much meaning as possible from less structured information sources must be devised. These methods include relational modeling of known metadata, such as To, From, Date, and Subject fields, flexible modeling of metadata provided through XML-based mechanisms, and deduction of additional metadata through semantic crawling and classification, and text mining. The ultimate information grid combines techniques for holistic data management, metadata management, and metadata deduction to extract full potential from all Oracle Grid Computing Page 9 enterprise information sources and easily exploit new information relationships as conditions change. IMPLEMENTING ORACLE GRID COMPUTING As more and more companies deploy clusters of industry standard servers and service-oriented software architectures, IT infrastructures resembling enterprise grids will naturally result. Oracle 10g is the first infrastructure software designed for enterprise grid computing and will provide a smooth transition from current IT infrastructure to a time-and-money saving grid computing infrastructure. IT departments looking to move to grid computing need to focus on three basic tasks: Consolidation of hardware, applications, and information shared among one or more data centers Standardization of servers, storage, and operating systems; use of common infrastructure services such as provisioning and identity management; standardization of application services made available as Web services; standardization of information sources and metadata Automation of all day-to-day management tasks, enabling a single administrator to simultaneously handle hundreds of servers in clusters Consolidation The first step toward grid computing is to consolidate your infrastructure, applications, and information in one or a few data centers. By consolidating, you can have fewer grids with larger pools of resources. The larger grids also increase reliability and reduce management costs. Oracle 10g is the only infrastructure that has full grid server cluster capabilities for all applications—transaction processing, decision support, and enterprise content management. That means that only Oracle can run your existing applications in a grid computing environment with no rewrite required. Standardization Every business today has a mix of server and storage technologies, applications, and information sources. An important part of lowering costs and increasing flexibility with grid computing is to designate standard units of hardware, application services, and information that will form the basis of a new grid. Of course, IT departments cannot and do not want to limit themselves to only a few standard components. However, the more resources that can be standardized, the lower the cost and the greater the flexibility of the overall infrastructure. Your grid can start with a single small cluster of servers and storage running one application. As you decommission old systems, you can add new capacity into your grid and move more applications over. Over time, the complexity of your infrastructure will drop and the cost savings will multiply. Oracle Grid Computing Page 10 Automation The third task is to automate your grid so that you can manage it effectively as it grows. Because enterprise grids can have hundreds of servers and thousands of application services, a grid is simply too large to be managed manually. Oracle Enterprise Manager 10g has automated the day-to-day maintenance required for an enterprise grid by providing a centralized management console called Oracle Grid Control. Software installation, patching, upgrading, workload balancing, security, and much more are all handled centrally from Oracle Grid Control. This means the entire infrastructure can be managed as one large computing system. One or a few administrators can maintain even the largest grid data center. HOW ORACLE USES GRID COMPUTING Oracle Corporation itself is using grid computing to improve its business processes, making our infrastructure more efficient and less expensive. Oracle’s Application Demo Systems (ADS) is used by its global sales organization to give demonstrations of Oracle products to prospects, customers, and partners. ADS runs the entire Oracle E-Business Suite—some 180 modules—in a single instance, with 450 copies of the environment maintained and supported for the sales force. In addition, we have a separate farm for technology demos that consists of about 50 environments. ADS hardware includes 300 Dell PE2650 servers running Linux, with dual P4 Xeon processors and 6GB of memory and 75 TB of storage. Running our demonstration environment—where performance is critical to a successful sale—on a Linux grid is a strong testament to the power and benefits of such an architecture. Linux on low-cost, commodity hardware is a proven technology for mission-critical applications such as Oracle E-Business Suite. In fact, Oracle runs its own business on E-Business Suite on a Linux/Intel middle tier. This deployment includes a Global Single Instance (GSI) in our Austin, Texas Data Center which provides mission-critical applications to more than 50,000 Oracle employees worldwide. The GSI database consists of Oracle9i Database with Real Application Clusters (3nodes) running on 3 Sun F12K machines with 104GB RAM and more than 3TB of storage; separately, the middle tier includes 21 Dell 2650 machines and 6GB RAM running Linux. The benefit to Oracle? Lower cost and faster performance. Running our EBusiness Suite internally on the GSI on Linux is 5 times cheaper and 3 times faster than on conventional UNIX/RISC. CONCLUSION Only Oracle’s grid computing architecture provides short-term and long-term benefits. With Oracle 10g infrastructure software, enterprise grid computing is available today. This is not a massive paradigm shift that will take years, millions of dollars, or an army of consultants to implement. Real businesses can begin to adopt grid technologies with minimal investment, zero disruption, and fast ROI. And Oracle Grid Computing Page 11 Oracle 10g software puts you on the path to even greater long term benefits as grid computing evolves and matures. Next Step For more information about Oracle grid computing, visit www.oracle.com/grid or call (800) 633-0615 REFERENCES  Forrester, 2004. www.forrester.com/go?docid=34449  IDC, 2004. www.oracle.com/technology/tech/grid/collateral/idc_oracle10g.pdf  Mainstay Partners ROI Series, www.oracle.com/customers/studies/roi  W3C, 2004, www.w3c.org/2001/sw Web Services and Service Oriented Architectures: A Savvy Manager’s Guide, Douglas K Barry, Elsevier Science, 2003. Loosely Coupled: The Missing Pieces of Web Services, Doug Kaye, RDS Press, 2003. “Grid’s Place in the Service Oriented Architecture,” Ian Foster, November 2004. “What is Service Oriented Architecture,” Hao He, April 2004. Oracle Grid Computing Page 12 Oracle Grid Computing February 2005 Author: Robert G. Shimp Contributing Authors: Miranda Nash Oracle Corporation World Headquarters 500 Oracle Parkway Redwood Shores, CA 94065 U.S.A. Worldwide Inquiries: Phone: +1.650.506.7000 Fax: +1.650.506.7200 www.oracle.com Copyright © 2003, Oracle. All rights reserved. This document is provided for information purposes only and the contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject to any other warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or fitness for a particular purpose. We specifically disclaim any liability with respect to this document and no contractual obligations are formed either directly or indirectly by this document. This document may not be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose, without our prior written permission. Oracle is a registered trademark of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners.
Pages to are hidden for
"Grid_Computing"Please download to view full document