Blog Post

A Window on the Cloud

Outsourcing compute power is wonderful — until something goes wrong. Unfortunately, when an Amazon Web Service goes down it’s hard to know why, and it’s even harder to know how well a particular cloud is performing in the first place. To make the cloud more transparent, open source cloud management software vendor Hyperic has launched, a web site that lets a user peek in on the various compute clouds to see how things are running.

CloudStatus measures service availability, latency and throughput for cloud-based infrastructure and application services. The initial release provides metrics for Amazon’s Elastic Compute Cloud, Simple Storage Service, SimpleDB, Simple Queue Service and Flexible Payment Service.

Hyperic sends a software agent to make requests against various cloud services, and according to CEO Javier Soltero, it racks up quite a large bill doing do. The web site views are free, but Soltero says Hyperic also plans to launch a line of services for paying customers. It’s a decent idea, but my worry is that Amazon or another cloud provider could shut the service down, either by offering their own status service or by stopping the Hyperic agent. Given the rush to provide dashboards, application-testing products and other services on top of established computing services, I’m eager to see how startups keep their footing in the clouds.

8 Responses to “A Window on the Cloud”

  1. If the cloud is spread across multiple geographies and data centers what exactly does the status mean? Is it the status of a specific data center?

    As Stacey points out, shouldn’t the cloud infrastructure provider display the status by default? If we consider the cloud to be a provider of OS like capabilities on the internet, we could expect the cloud to provide status as well.

    Hyperic is providing the equivalent of Systems management tool in the cloud. It will be interesting to see the evolution of Systems Management for the cloud.

  2. I am (or was, maybe still am) in the camp that thinks my mid-tier clients are better of in the cloud for CRM services. some types of native database applications, POS, etc.

    But, some of these venture backed SAAS and PAAS providers simply cannot warrant that they will be around in perpetuity. How can they?

    The industry needs to underwrite some type of continuity indemnification, insurance standards beyond vanilla business interruption, and potential solutions for hot standbys.

    It’s going to get ugly after the first high profile SAAS providers go into Arrington’s dead pool, or a service like Amazon or Google has a big, costly outage. Then people will re-think what using VC backed cloud services can buy. Still it’s tempting and a sexy alternative for those who can plan around the inevitable outages.