HP 3000 Manuals HP 3000 Manuals
Balancing Workload During System Consolidation The Workload Manager can facilitate the process of consolidating multiple source systems onto a target system. You can use workgroups to help you plan and execute the consolidation, and to manage the final consolidated system. To partition the workload on the target system One typical concern regarding consolidation is the limited amount of control the system manager has over the target system. Before the introduction of the Workload Manager, five scheduling subqueues were available on each of the source systems, and only five scheduling subqueues were available on the target system. As a result, the scheduling subqueues on the consolidated system contain a larger numer and varieity of processes than did the source systems, which decreased the system manager's ability to control CPU scheduling. With the introduction of the Workload Manager, you can now define multiple workgroups to represent the users of each of the source systems. For example, suppose you are consolidating three systems. In this case, you could create workgroups to represent the CS, DS, and ES processes from each one, for a total of nine workgroups on the target system. This preserves the partitioning that had been available with the physical separation of the source systems. Alternatively, you can use the Workload Manager to define workgroups that more naturally reflect the needs of the combined user population. Perhaps data entry clerks had been in the CS subqueue of several source systems and now you can combine them into a single workgroup on the target system. You might collect similar batch jobs into a common workgroup. Or, you could separate users who were once forced to share the CS subqueue into distinct workgroups. The scheduling characteristics of the workgroups on the target system can be adjusted to result in the CPU access that the system manager requires to achieve desired performance. To manage user expectations Another area of concern relates to the effect of the consolidation itself on user's experience and expectations of system performance. Consider a situation where you are consolidating Systems A, B, and C onto System D over a period of time. You plan to bring over System A on the first weekend, System B on the second weekend, and System C on the final weekend. During the first week on the target system, while the first set of users is running alone on System D, the response time and throughput are excellent. When they are joined by the users from System B, their performance may degrade. Once all three systems are combined on the target, the System A users may actually complain about their performance. Even though it may be better than what they had on System A, the performance degraded as additional users were added to System D. How can the Workload Manager be used to help this situation? The first set of users on the target system grew dissatisfied because they had become accustomed to the better performance when they had exclusive use of the target system. To solve this problem, you can restrict the amount of CPU available to users. For example, constrain System A users to 30% of the target system so that they will experience, from the onset, the performance that will result when the entire consolidation is complete. This example is obviously simplified. You may not wish to divide the target system up evenly among the users from the three source systems. Perhaps one set of users is more important and requires more of the CPU. Alternatively, the consolidation may raise the larger concern of how to ensure satisfactory coexistence of competing workloads from the various source systems once they have been consolidated. The Workload Manager gives you the tools you need to create an effective workgroup configuration, monitor the performance, and make adjustments as necessary to ease the consolidation process.