JobServer Overview

• Need to manage thousands of jobs with peak load requiring hundreds of jobs be run concurrently.
• Have only a few dozen jobs to manage, but each job gets scheduled hundreds of times during the day.
• Have jobs with sensitive response times, so a job needs to get kicked off with a high degree of precision.
• Jobs that run for only seconds and some jobs that run for hours or days.
• Have a large number of users that need to edit, create, track, and administer jobs with fine grained privileges over the various tools and job groups.
• Have high-end hardware with lots of processing power that needs to be utilized.
• Complex scheduling rules.
• Need for a rich set of tools for managing jobs that also allows for a large number of concurrent users with varying access rights and privileges.

JobServer offers a highly reliable, scalable, and manageable foundation from which to base your company's job processing environment. Whether you have jobs that are geared toward internal data warehousing, ERP processing, accounting, batch processing, monitoring, or B2B transactions, JobServer might be what you are missing. These jobs maybe internally IT facing and/or externally customer facing. JobServer is a solid foundation to base your processing needs upon. Through JobServer's scalability and powerful APIs, you will have the tools and the power you need. For a detailed list of all the features available in JobServer go here.

JobServer comes with a rich set of GUI tools by which to manage your jobs and environment. You can design, schedule, run, and track your jobs with a high degree of precision and detail. Know how, what, and when a job ran. Know who last modified a job's schedule, who last shutdown the server ...etc. If a job throws errors or warnings get notified by email or quickly search for such conditions across any set of jobs (great reporting features available). When a job runs, know exactly what was sent to the customer. Know what inputs the job used when it ran and what outputs it generated. JobServer knows everything about a job from the time it is created all the way through each and every run.

This is only a sample of what JobServer is capable of. The GUI tools available with JobServer are network enabled requiring no software to be installed on the client end. The tools can be accessed with standard web browsers and through Web Services. With adequate hardware, the JobServer GUI tools can support a large number of concurrent users. Users can be granted restricted rights and privileges to certain tools. For example, you can grant one user the right to view a job's schedule but not to edit it. Users can also be granted rights to certain job's and job groups allowing for fine grained access control.

The JobServer Scheduling Engine leaves no scheduling rule out of your reach. Virtually, any repeating or one-off scheduling pattern that you can imagine can be represented using the JobServer Scheduling Engine. As an example, here is a sample of scheduling rules that are supported:

• Every 15 minutes starting on 1/1/2011 09:00AM and only between 10:00AM and 7:30PM and ending on 3/3/2012 5:00PM
• Every other month starting on 1/1/2011 09:00AM and on the last, first and second Monday and last Friday of the Month
• Every other Week on Monday and Wednesday starting on 7/7/2011 except for Wednesday of 7/30/2011
• Many more...........

These examples only scratch the surface of the scheduling rules and options available. Other features include:

• Multiple one-off dates can be added or excluded to any schedule. For example, this is a convenient way to include or exclude company holidays.
• Multiple intra-day time ranges can be excluded (e.g. exclude 1:00am to 4:00am)
• If schedule times are missed due to down time, past scheduling times can be skipped ahead to the future or can be executed one at a time until the schedule has moved to a date in the future.
• Jobs can be set to stop running after a certain number of scheduled executions (e.g. stop after the job has run 55 times).
• Expire window requires a job run within a certain time window from its scheduled time, otherwise, it will "expire" (administrators can be alerted when this happens).
• Multiple levels of email alert can be setup when jobs fail. This can be setup at the entire JobServer level, Partition level, Job Group level, and/or individual job level.
• Many more......

Keep in mind also that a job can be manually launched by a user at any time and as often as needed without being scheduled. So if a job needs to be rerun for any reason, it can be rescheduled or manually run by the user as necessary.

The JobServer Scheduling Engine makes sure that jobs run when they are supposed to. If a job can't run (e.g. system failures), then the engine will keep trying to run the Job. If the Job fails while running, the status of the job can easily be tracked to allow the administrator to diagnose any problems. The engine is also highly scalable and can manage the scheduling chores for thousands of jobs.

JobServer can scale with the hardware that is available to it. It is designed to take advantage of multi-CPU and multi-core systems. The JobServer uses an advanced look-ahead caching scheme to make sure that no time is wasted when preparing to run a Job. It employs an advanced multi-thread scanning design to efficiently detect what Jobs are ready to run and quickly prepare them for running. The Scheduling Engine and corresponding job processing unit can be segmented into any number of Partitions. Partitions allow for higher degrees of concurrency and are vital for installations that have a large number of jobs and where scheduler promptness is critical. A Partition can also be defined to limit the number of concurrently running Jobs to avoid resource overload. Whether you are manging a handful of jobs or tens of thousands of jobs, JobServer is up the task.

Job dependency rules can be defined to control when jobs can run relative to each other. For example, say you have job A that must run and complete each day before a group of other jobs may be allowed to start. In this case, the jobs in the group will not be scheduled so long as job A has not completed and satisfied the dependency.

Imagine you have a job that sends an email to a customer. Now next month you know that you will need to change this email address. Job Time Versioning allows you to change a job's properties and have them take effect at a future time. You may set an arbitrary number of future versions for a job.

Segment your JobServer environment into multiple computing Partitions to manage and control CPU, database, and other resource utilization. Further organize jobs into Groups and then restrict users rights to only the Job Groups and Partitions you want them to have access to.

JobServer goes beyond just general reporting of when jobs run and how they completed their processing. Know exactly what inputs the job used when it was run and exactly what outputs it generated as it ran. JobServer can track the exact inputs/outputs used by a job for each and every run. This give the user and administrator the ability to go back into history, at the exact time the job was run, and know how it was configured. This allows you to give customers and users detailed proof of what was or wasn't done to the job.

The goal of the developer API is to allow companies to build their own jobs in a modular and reusable fashion as they so desire. Many of the features are optional, so you may choose the level of functionality and reusability that you wish to employ. To create a job that will run in JobServer, all it takes is implementing one interface and putting the developed Java software into an appropriate JAR file. The job can be one line of code or thousands of lines of code. If you are interested in more advanced reusability and extensibility features then continue reading below. JobServer also comes prepackaged with built-in components called SOAFaces that can, for example, execute unix bash, ksh, and perl scripts and others that can send email alerts, FTP files, make Web Services calls, or perform XML processing.

The SOAFaces API is an innovative foundation that JobServer uses to define what a job does and how it executes within a server. The API is developed on an open source foundation allowing developers to get involved and drive innovation even further, and without vendor lock in. There are also a sub project called BeanSoup that builds on the SOAFaces. BeanSoup is a collection of reusable SOAFaces components implemented and contributed by a community of developers. Even if you are not using JobServer get involved!

We have all heard about the promise of component software. Buy/Build software and connect it together to compose more complex functions. Well, JobServer is driving innovation by allowing developers to build and compose component software called SOAFaces. Every job that is scheduled and executed within the JobServer environment is composed of one or more SOAFaces. A SOAFaces component (aka tasklet) can be as simple as a single line of code or as complex as ten thousand or more. Developers have the option of developing these tasklets in such as way that they can accept input and send output to other tasklets. This is all done using a simple POJO (plain old Java objects) API, so that there is no vendor lock in or complex API for developers to learn. All the heavy lifting is done by the JobServer Container. As an example, all logging is done using the standard Java Logging API where the JobServer Container manages the log output. GUI tools are provided to search and view the logs. If a tasklet wishes to expose its outputs to other tasklets, it can do so by defining a simple JavaBean and it can optionally define what inputs can be passed to it from other tasklets (using an output JavaBean). What's more, is that users can also visually edit these input properties on the JavaBean using JavaBean GUI Customizers. Most of the APIs and features are optional so the developer can chose the level complexity they want. For more details read here.