Identify Infrastructure Qualities
We have already covered infrastructure qualities in objective 2.3. To recall them, here is a short overview:
Availability is the ability of a system or service to perform its required function when required. It is usually calculated as a percentage like 99,9%.
Manageability describes the expense of running the system. If you have a huge platform that is managed by a tiny team the operational costs are very low.
Performance is the measure of what is delivered by a system. This accomplishment is usually measured against known standards of speed completeness and speed.
Recoverability describes the ability to return a system or service to a working state. This is usually required after a system failure and repair.
Security is the process of ensuring that services are used in an appropriate way.
Key Performance Indicators
According to ITIL, a Key Performance Indicator (KPI) is used to assess if a defined service is running according to expectations. The exact definition of the KPIs differs depending on the area. This objective is about server performance which is typically assessed using the following KPIs: Processor, Memory, Disk, and Network. VMware offers several concepts for managing resources.
To manage CPU resources VMware relies on the the CPU scheduler. The CPU scheduler shares the same logical processor among multiple virtual machines. It defines the following terms:
- Processor Socket: A physical CPU
- Core: A logical core within a physical CPU
- Logical Processor: A hyperthreading CPU Core presents itself as multiple logical processors
VMware offers the following features to manage the memory efficiently
- Transparent Page Sharing: Shares identical Memory Pages among multiple virtual machines. This feature is active by default and does not impact the performance of the virtual machine.
- Ballooning: Controls a balloon driver which is running inside each virtual machine. When the physical host runs out of memory it instructs the driver to inflate by allocating inactive physical pages. The ESX host can uses these pages to fulfill the demand from other virtual machines.
- Memory Compression: Prior to swap memory pages out to physical disks the ESX server starts to compress pages. Compared to swapping, compression can improve the overall performance in an memory overcommitment scenario.
- Swapping: As the last choice the ESX hypervisor starts to swap pages out to physical disks. This is definitely a bad situation as disk are much slower than memory.
Storage I/O Control (SIOC) allows cluster wide control of disk resources. The main goal is to prevent a single VM to use all available disk performance from a shared storage. With SIOC a virtual machine can be assigned a priority when contention arises on a defined datastore.
Network I/O Control (NetIOC) enables traffic prioritization by partitioning of network bandwidth among the entire cluster.
VCAP5-DCD Exam Blueprint v1.1
- Understand what logical performance services are provided by VMware solutions.
- Identify and differentiate infrastructure qualities (Availability, Manageability, Performance, Recoverability, Security)
- List the key performance indicators for resource utilization.
Skills and Abilities
- Analyze current performance, identify and address gaps when building the logical design.
- Using a conceptual design, create a logical design that meets performance requirements.
- Identify performance-related functional requirements based on given non-functional requirements and service dependencies.
- Define capacity management practices and create a capacity plan.
- Incorporate scalability requirements into the logical design.
- Determine performance component of SLAs and service level management processes.
Back to VCAP5-DCD Study Guide