【正文】 ational batch activities can begin, the basic ponents that make up the batch architecture should be understood. A batch architecture consists of the processes and ponents used to effectively manage batch processing. This section is a brief introduction to the typical ponents found in batch architecture. The purpose of the batch architecture is to optimize processing (improve response time and utilization of system resources) by executing batch runs during offpeak periods. The architecture should provide the capacity manager with an easy to use interface and permit a standard and centralized approach to batch scheduling, monitoring, control, and error correction. The architecture should be highly scalable in order to meet the future needs of the anization. It should also be highly available, with minimal downtime, and minimize impact to online operations, which usually are operating concurrently with the batch operations. Some anizations may decide to have backup ponents, such as the event server, to ensure the pletion of all missioncritical batch jobs. Activ ities Training Start Batch Architecture Batch Processing Job Scheduling Documentation and End Service Management Function 9 Figure 2 shows the basic ponents of the batch architecture, which include the management server, capacity database (CDB), monitor, printer, application servers, and databases. In addition to the monitor attached to the management server, each application server should have a monitor to permit viewing of local batchprocessing activity。 some scheduling tools may have the ability to attempt to restart the batch run. If the initialization is successful, the first batch job is executed. The scheduler manages the execution of jobs on each application server targeted to perform the batch processing. If no errors are encountered during the execution of the job run, then the tool records the successful pletion of the job in the batch log and the next job is executed, and so on. If an error occurs during the execution of a batch job, the scheduling tool stops processing that job and an error is sent to the error log. The actual execution of a batch job is dependent on a number of inputs that could be the reason that a job is not executed. For example, the following inputs may be required: ? Availability of production data ? Completion of a job on which the queued job is dependent ? Availability of resources to execute the job ? Priority of the job and the batch window Warnings may also be generated during job processing if, for example, CPU or disk space capacity exceeds the threshold value for an application server processing the batch job. Warnings should not stop a batch job from being pleted。 jobspecific information should be coded into a designated area within the skeleton. The skeleton also serves to facilitate development and maintenance requirements by standardizing the content and structure of each job script. For example, any type of standardized actions, such as the notification of a successful or nonsuccessful batch job execution and transaction data archiving and deletion, should be included in the code of every script that is executed. Scheduled batch runs are initiated by a scheduling tool during a predefined batch window, typically when user activity of the system is at a minimum. After the scheduling tool has been programmed, capacity manager interaction should not be necessary during batch runs unless an error, from which the tool cannot recover, occurs. Figure 4 is a flow chart of a typical scheduled batch process. This highlevel overview is meant to give the reader a general understanding of how scheduled batch jobs are processed. Keep in mind that the capabilities of the scheduling tool determine exactly what functionality can be automated. Figure 4. Batch processing flow chart Run Success No Yes Run Halts Batch Scheduler Log File Recorded in Error Batch Job A Execute Job Halts Batch Scheduler Fatal Warning Fatal or Yes Error Warning No (etc.) Batch Job B Execute Scheduler Initiates Batch 14 Job Scheduling The scheduling tool initiates all batch runs. If the run does not begin as scheduled, the tool halts the run and records an error in the error log。 Job Scheduling Service Management Function Published: October 20xx Reformatted: January 20xx For the latest information, please see ii Job Scheduling The information contained in this document represents the current view of Microsoft C orporation on the issues discussed as of the date of publication. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a mitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information presented after the date of publication. This document is for informational purposes only. MICROSOFT MAKES NO WA RRANTIES, EXPRESS, IMPLIED O R STATUTORY, A S TO THE INFORMATION IN THIS DOCUMENT. C omplying with all applicable copyright laws is the responsibility of the user. Without limiting the rights under copyright, this document may be reproduced, stored in or introduced into a retrieval system, or transmitted in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), but only for the purposes provided in the express written permission of Microsoft Corporation. Microsoft may have patents, patent applications, trademarks, copyrights, or other intellectual property rights covering subject matter in this document. Except as expressly provided in any written license agreement from Microsoft, the furnishing of this document does not give you any license to these patents, trademarks, copyrights, or other intellectual property. Unless otherwise noted, the example pa