Introducing the Workload Concept
|WorkloadCentral broadly defines a workload as the I/O requests made by applications against storage infrastructure. When these requests are analyzed in terms of IOPs, throughput, read or write percentages, metadataa ratios, sequential or random behavior, block size distribution their characteristics can be used as the basis for modeling that Workload. The resulting model can be used to emulate the Workload to support numerous activities across the storage lifecycle including development, pre-production or production for troubleshooting, testing and validation, technology and product evaluation, capacity planning, performance analysis and optimization.|
At center of WorkloadCentral is the Workload Library which is a community resource to share and download different workloads. There are three different categories of workloads available within the Library each available from the respective button at the top of the Workload Library page.
- Workload Templates are base workload models developed by Virtual Instruments (formerly Load DynamiX). They are designed to be copied (known as cloning) and configured to specific user workload modelling needs.
- Example Workloads provide example application workloads based on publically available research on different application workload behavior.
- Community Workloads are workloads created by WorkloadCentral community members either by copying (cloning) existing workloads (including templates and examples) or from based on analysis of their workload data.
- Protocol Workloads: provides out-of-the-box Protocol Workload models for key file, block and object protocols that make it easy to create and model storage Workloads specific to your environment or applications. A Protocol Workload is not based on any specific application, but a template of protocol parameters and I/O characterization that allows you to customize to emulate an application workload. In a Protocol Workload Model, key workload parameters can be varied such as % of read vs % of writes, % random vs % sequential, distribution of file or block sizes, etc.
- Constant-Load Workloads: are Protocol Workloads that generate constant Load (i.e. IOPS and Throughput) throughout the duration of the Workload. Constant-Load Workloads are recommended for those who understand the characterization of their production workloads, but does not need to vary the Load over time.
- Temporal Workloads: are Protocol Workloads generate Load (i.e. IOPS and Throughput) that varies over time according to user definition throughout the duration of the Workload. Temporal Workloads are recommended for those who want to understand production workloads characteristics over time. This type of workload is data driven and can only be created by importing your data into WorkloadCentral.
- Application Workloads: Application Workloads differ from Protocol Workloads in that Application Workloads provide parameters that are specific to the Application, such as Number of VMs and Linked Clone Size for VDI Application Workloads, instead of protocol parameters such as % of read vs % of writes in the case of Protocol Workloads.
- TDE Workloads: are specific Protocol Workload Models or Application Workloads authored in the Virtual Instruments (previously Load DynamiX) Test Development (TDE) environment. TDE Workloads are recommended for those who are storage protocol experts who are looking to validate or test specific protocol level functionality, capacity or performance.
Workload Use cases across the Storage Lifecycle
Workload Models can be used across the storage lifecycle in support of key functions or activities like:
- Technology Evaluation: which storage architecture, technology or protocol offers the highest performance or is the most cost-effective when running workloads that reflect your specific applications.
- Product Evaluation: which vendor and storage system offers the highest performance or is the most cost-effective when running workloads that reflect your specific applications.
- Configuration Optimization: what is the optimal storage price / performance configuration by varying dozens of settings like load, block/file size, compression/de-duplication, tiering, queue depths, and caching.
- Problem resolution: analyze production workloads performance for transient issues or other systemic issues, and quickly reproduce the workload behaviors in the lab for faster resolution.
- Change Validation: test, verify and validate infrastructure configuration and firmware changes, application upgrades and their effect on scalability, latency, throughput of storage infrastructure before cutting over to production.
For more information on workloads and WorkloadCentral see the following sections: