Information Technology Reference
In-Depth Information
VEs as cluster nodes improves isolation between workloads, making it easier to
run more applications in the cluster.
1.1.6.2 Asynchronous Workloads
Many workloads are active for a certain period of time each workday but other-
wise use few system resources. For example, employees of a small business may all
work in one time zone, and their online transaction processing (OLTP) environ-
ment might have few transactions running at night. Conversely, a batch job may
run for 6 hours each night. Although the business may use two separate systems
for these workloads, the average utilization for each system will be less than 40%.
You can combine those two workloads on one system in several w a y s . Fo r ex -
ample, you can take advantage of their schedules by running both workloads on
the same system, hoping that there will not be any adverse interactions between
them. Alternatively, you can prevent adverse interactions by isolating the work-
loads with virtualization. If the two workloads need roughly equivalent resources
to provide the desired response time and completion time, the graph of utilization
versus time might look like Figure 1.8.
Figure 1.8 CPU Utilization of Asynchronous Workloads
1.1.6.3 Software Development and Other Bursty Workloads
A software developer typically generates intense system activity during short pe-
riods of time when compiling software, followed by periods of little activity. This
uneven workload results in computers that are under-utilized most of the time,
wasting most of the investment in those systems.
 
Search WWH ::




Custom Search