What is an application workload?
The footprint of an application as it consumes computing resources in terms of CPU, memory, I/O, and network. A defined workload can be evaluated in terms of performance (how easily the computer handles the workload), which is generally defined by response time (the time between a user request and a response to the request from the system) and throughput (how much work is accomplished over a period of time). Because computing requirements are varied, so are the workloads.
Types of application workloads; some of the typical application workloads you might find in a hybrid cloud environment include
- Batch workloads, which tend to process huge volumes of data
- Transactional workloads which often are the automation of business processes such as billing and order processing
- Analytic workloads, where the emphasis is placed on the ability to holistically analyze the data embedded in these workloads across public websites, private clouds, and the data warehouse
- High-performance workloads, usually specialized processes with scientific or technical requirements
- Database workloads, which are an extremely common type of workload, which must be tuned and managed to support the service that’s using that data.
Why monitoring application workloads is important
- Application Performance Monitoring tools monitor business transactions as they move through a custom application written using Java or .Net and establish dynamic baselines. These tools can track every line of code and initiate deep diagnostics if performance wavers. Many can also monitor servers (CPU & disk utilization, memory consumption) and databases (performance metrics like resource consumption, database objects, schema statistics). Customers can focus on revenue and conversion rates rather than focus on application performance issues alone.
- However, if you have 1000 applications you might license and instrument APM to monitor 10% of them – only the business-critical applications. For budgetary reasons, you wouldn’t have APM instrumented for the remaining 90% of applications which may be tier 2 and tier 3 and may include commercial off-the-shelf applications like Backup or Authentication where code-level visibility is not of much value. While your application monitoring tools monitor applications, virtual servers, and databases, they can’t see the underlying SAN or storage infrastructure. However, this shared SAN infrastructure could be used by a noisy neighbor application which adversely impacts the SLA of a tier 0/1 application.
An infrastructure monitoring platform like VirtualWisdom monitors the entire stack – from the host or VM down to the HBA port, SAN fabric and ports on your networked storage array.