SHARE

ZeroStack added new monitoring and alerts to give companies more visibility into its cloud platform.

The platform is essentially a private cloud-in-a-box that can be consumed as if it were a public cloud, but on premises. It’s a managed service running ZeroStack’s software stack. Customers choose their own hardware.

These customers are typically mid-market sized companies with between 10 and 20 IT professionals — businesses that can’t hire and maintain a large cloud team. Steve Garrison, VP of marketing at ZeroStack, won’t say how many companies use the cloud platform, only that it numbers “well past double digits” and includes cybersecurity firm Zimperium.

About a year ago the company added machine learning capabilities to the cloud platform.

This week it added customized alerts and thresholds.“This augments our machine learning,” Garrison said.“We’re giving people additional visibility and allowing them to create their own parameters.” Alerting policies can be defined on specific host-based CPU utilization, storage, network bandwidth, and other parameters.

“Let’s say your OpenShift was running fast yesterday and it’s slow today,” he explained.“With our new thresholds we can see that the CPU utilization was driven up by three other business units and that might be degrading its performance.”

The new alerts and thresholds also allow IT administrators to set multiple policies on one or more physical hosts to monitor CPU utilization, memory utilization, network rates, and storage utilization and storage rates, so they have fine-grained alerting capabilities across multiple resources. Alerts can be routed through various notification channels including email, PagerDuty, and ServiceNow.

These metrics enable better IT management, Garrison said. “It really is a performance management play on top of cloud infrastructure. Cloud is cool, but there’s a lot of abstraction and a lot of confusion because of that. You want more triggers and ways to create alerts; we’ll give it to you. You can’t manage what you can’t measure.”

<<< This article was originally published on SDxCentral’s website here. >>>