What is a cloud computing ‘storm’ — if such a term exists… and what are its implications for software application developers working in newly virtualised platform environments?
Also sometimes known as ‘boot storms’ or VDI boot storms, a storm is a degradation of service experienced when a large number of users boot up a cloud service inside a narrow time frame.
The effect of the storm is that the network becomes overwhelmed with data requests — and so, logically, performance naturally suffers.
According to TechTarget, storms are a problem because slow, unresponsive virtual desktops can make a VDI project unpopular with end users even though desktop performance is good the rest of the time.
“To prevent VDI boot storms, experts recommend starting virtual machines (VMs) in waves, implementing caching or adding solid-state storage to handle the concentrated number of storage input/output (I/O) requests,” reads the official definition.
Naturally then, the IT industry has been busy developing layers to help combat the effects of cloud computing storms.
Hybrid-cloud performance management company Xangati’s own Xangati Virtual Appliance (XVA) architecture features automated storm remediation for virtualised and VDI infrastructures.
According to the firm, even the best run cloud environments are hit by storms of all kinds such as storage, CPU, memory and boot storms, which cause contentions that cripple applications and disrupt end-user experiences.
Virtualisation System Administrators (V-SysAdmins)
With this latest Xangati release, virtualisation system administrators are able to remediate CPU and memory performance issues by automatically balancing workloads across vCenter hosts.
“Xangati is moving closer to its long-term vision of an autonomic infrastructure management that delivers an automated way to stave off complex degrading conditions and v-storm contentions, in addition to end-to-end, closed-loop orchestrated visibility to real-time performance indicators,” said Atchison Frazer, VP of marketing at Xangati.