Mike Kiev - Fotolia
As computing power moves out of traditional data centers and migrates toward the edge, virtualization must move as well, but edge computing poses unique challenges for virtualization. The edge lacks many of the general-purpose, enterprise-class servers with high-end processors that virtualized environments typically use, and instead relies on purpose-built devices such as IoT sensors, which feed busy network traffic to just a few edge servers. Performance demands at the edge often require processors built with established reduced instruction set computing architectures such as Advanced RISC Machine.
For virtualization to fully migrate to the edge, hypervisors must embrace and function on reduced instruction set computing (RISC) architectures. VMware offers early stages of such support and has already demonstrated ESXi running on Advanced RISC Machine (ARM) processors.
RISC architectures, such as ARM processor architecture, aren't essential for virtualization. However, the way organizations use such processors matters when implementing virtualization in edge environments.
CISC vs. RISC processors
People generally categorize the processors that servers and PC-class systems use as complex instruction set computing (CISC) processors. These processors support full instruction sets and powerful features, such as multi-threaded instruction pipelines, intended to enhance performance.
However, capability means complexity. High-end CISC processors can contain hundreds of millions of transistors, each contributing to the processor's power usage. This sometimes yields mediocre performance for certain tasks simply because of the unavoidable latencies that a CISC processor's complexity can create. Hypervisors were designed to use CISC processors, which is why such processors work incredibly well when implementing virtualization in most environments.
However, simple or more dedicated compute platforms sometimes waste the versatility and command/feature set of CISC processors. RISC processors, by contrast, are simple. By removing support for extraneous commands and features, chipmakers can vastly reduce the number of transistors in the processor, which boosts performance while reducing the processor's power consumption.
Virtualization at the edge
Although you can use a few CISC servers to collect and process edge data, you might find arrays of dedicated appliances, such as network and security equipment, and vast IoT device deployments more useful. Each IoT device functions as a dedicated computer that often operates on low battery power and runs only a limited software base to operate and manage that IoT device. Consequently, dedicated appliances and IoT devices rarely require the varied capabilities of a CISC processor.
Virtualization at the edge can provide versatility and management flexibility to a hardware environment. If you can virtualize a RISC processor, your dedicated device can then interoperate with the broader virtualized infrastructure. Virtualization tools can "see" virtualized appliances and IoT devices and apply the same monitoring and management functionality that traditional enterprise IT uses to both.
However, most hypervisors can't virtualize RISC processors while remaining compatible with mature hypervisors already in service on CISC platforms.
Dig Deeper on Selecting storage and hardware for VMware environments
Related Q&A from Stephen J. Bigelow
Just because software passes functional tests doesn't mean it works. Dig into stress, load, endurance and other performance tests, and their ... Continue Reading
Don't neglect form factor as part of your data center server selection. Instead, figure out what type of environment you need and learn which server ... Continue Reading
Learn how load balancing in the cloud differs from a traditional network traffic distribution, and explore the different services available from AWS,... Continue Reading