January 2012
Focus on Automation to Make Your Data Center More Efficient
The economy is slowly turning around and business is improving. Meanwhile, your data center is growing, your transaction count has increased, and you’re adding and upgrading equipment. But, your company is still reluctant to add more staff to meet your growing resource needs.
You probably have fewer resources, but more servers, processes, and data than ever. Begin this year by making your data center more efficient through automation. Here’s the secret to getting started…
Manage by Exception
What does “Manage By Exception” mean? Well, it basically means that your day is busy enough without being tied up monitoring routine tasks. Learn to trust automation to handle the routine. That way, you deal only with the exceptions.
When you first automate new processes, add notification milestones along the way that tell you when critical processes, like a database update or a nightly backup, have completed. Later, when you’re comfortable with the entire automated process, you can remove these notifications. Here are the keys to managing by exception:
Embrace Escalation
Make sure that each piece of automation you’ve put into place contains a link to the outside world (such as an SNMP connection to an enterprise monitor) to route important issues to the correct people.
For example, you might want to notify the Operations group, or the person on call, that the order extract from your web site failed, the data backup to your SAN ran into storage space issues, or a critical ACH job is late. The notification should route the issue to the proper expert, or group of experts, for faster resolution.
Monitor to Manage
Monitor critical system parameters that could affect your automation using an automated monitoring tool. Monitor disk space, system performance, memory management, and system messages. Don’t rely on an operator to manually review memory pools, disk usage, or monitor messages—you won’t get timely information. It’s critical to automatically and proactively monitor your system by setting notification thresholds before processes stop.
Don’t Scrimp On Security
Poor security can really compromise your automation. Secure your systems properly to prevent unauthorized system access. Encrypt data when necessary (on tape or disk, down to the field level), and create an audit trail for database and system-level operating system changes. Your security software should notify you when there are exceptions and escalate the message to the correct personnel.
By Chuck Losinski, Director of Automation Technology
Easy, Powerful, Enhanced Network Control with Robot/NETWORK 11!
Soon, Robot/NETWORK 11 will be available. This new release offers a number of useful enhancements, including an updated look and feel, and an easy-to-use, powerful Product Metrics dashboard, which reveals summarized automation information for your products—statuses, jobs, messages and monitoring, report sets, and save media.
We’ve also enhanced the Robot/NETWORK Status Center so you can assign status messages to a user. For example, you could assign a status message from Robot/SAVE that your backup failed to your night operator. And, there are new options to display your assigned statuses, or all assigned statuses. Plus, we’ve added more status escalation flexibility.
Updated Look and Feel
One of the first things you’ll notice when you start Robot/NETWORK 11 is the updated look of the Robot/NETWORK Explorer and its more intuitive design.
Figure 1. Updated Robot/NETWORK 11 Explorer
Product Metrics Dashboard
When you click the Product Metrics button in the Robot/NETWORK Explorer, a dashboard displays containing Robot/NETWORK, Robot/SCHEDULE, Robot/CONSOLE, Robot/REPORTS, and Robot/SAVE statistics.
Data is included from the past week, month, year, or all the history data collected. Once a day, each Robot/NETWORK Node collects and summarizes data for each Robot product that you own. Then, the Node transmits these summary records to the Host to display in the dashboard.
Figure 2. Robot/NETWORK Product Metrics Dashboard
If you want more specific information, click Details to display the Product Metrics Details window and specify a date range.
Figure 3. Product Metrics Details Window
Status Center
We also enhanced the Robot/NETWORK Status Center. Now, you can assign a QSYSOPR message or product status message to any user with a user profile on the Host system by right-clicking on the status.
Figure 4. Assigning A Status
When you display the status by clicking Properties…, you see details about the status, including the user assigned (you can assign a new user by clicking the Search button to the right of the Assigned To field).
Figure 5. Status Properties
When you assign a status, you also prevent further escalation of the status from Robot/NETWORK through SNMP or Robot/ALERT. And, you can add a ticket number and notes to the status.
Figure 6. Notes Tab and Fields
Escalation
Robot/NETWORK 11 can escalate statuses to the outside world with greater flexibility. It has separate timers for each type of status escalation: a Robot/ALERT message, an SNMP trap, or a user-defined program. You can specify one or more escalation options, either in series (one at a time), or simultaneously (all at once).
For example you might want to have Robot/ALERT send an e-mail to the Operations group after five minutes. Then, if there’s no response after 30 minutes, use SNMP to escalate the status to your enterprise monitoring software. At the same time you could have your program log the issue. Or, you could escalate the issue using all three methods at a certain time after the status occurred. It’s up to you.
That’s a quick overview of Robot/NETWORK 11. Look for more information coming your way when Robot/NETWORK 11 is ready for prime time early in 2012.
By Chuck Losinski, Director of Automation Technology
Q & A
Here are some recent questions we received about Robot product enhancements and their answers.
Robot/CONSOLE
Q: How much information do I receive and what action can I take based on my Robot/CONSOLE resource monitors?
A: We recently enhanced job resource monitoring to retrieve additional information from the job— job number, job user, job function, job type, current user, actual job status—which you can use with message variables in OPAL.
For example, you could end a job that’s using too much CPU (with an ENDJOB command in OPAL) and forward specific information about the job using Robot/ALERT.
Robot/SPACE
Q: I run multiple small disk collections. Is there a way to report on these collections?
A: Yes, now you can run the Good Morning Report for more than one collection at a time.
Robot/SCHEDULE
Q: Can I use the Robot/SCHEDULE Blueprint interactively with my job schedule?
A: Yes, you can perform job override options, such as DO and HOLD, by right clicking on a job within the Robot/SCHEDULE Blueprint.
Purchase any software solution from Help/Systems (Robot Automated Operations Solution); PowerTech (IBM i security solutions); SEQUEL Software (data access/analysis and productivity software); Bytware (anti-virus and monitoring solutions for IBM i) and enjoy big discounts on training, services, and IBM POWER7 systems. For details, contact your local IBM Business Partner, or visit our IBM i Solution Edition web page.