CloudPhysics helps IT run VMware like Google

"Google uses analysis of anonymized traffic data from everyone's GPS location streams to help users avoid accidents and bottlenecks and to make better driving decisions. CloudPhysics brings that same kind of power to IT so enterprises can make better operational decisions," said John Blumenthal, CloudPhysics CEO and Co-Founder. "Our servers receive a daily stream of 80+ billion samples of configuration, performance, failure and event data from our global user base with a total of 20+ trillion data points to date. This 'collective intelligence,' combined with CloudPhysics' patent-pending data center simulation and unique resource management techniques, empowers enterprise IT to drive Google-like operations excellence using actionable analytics from a large, relevant, continually refreshed dataset."

Virtualization Adds Complexity -- and Risk
Virtualized environments are larger, denser and more complex than ever before. More integrations with other systems in the IT stack now exist, involving subtle interdependencies. Consequently management of these environments has also become very complex, and remedial actions often have unintended negative consequences. As a result, admins often resort to shortcuts or simply ignore tedious management capabilities.
For example, as virtualization packs applications closer together, it sharply increases the number of operational hazards hidden in the interdependencies among network, storage, memory, cpu, availability and other systems. IT teams have zero visibility into this underlying and constantly changing complexity. They only discover hidden problems when they cross safety margins, triggering urgent, dangerous and suboptimal responses. The software-defined data center, built upon virtualization, only adds more abstractions which further complicate operations management.
Managing VMware High Availability (HA) environments exemplifies this. Adding a single server or changing the resource allocations of an existing virtual machine can dramatically reduce the safe consolidation capacity of the entire cluster. The inability to fully and safely utilize has hardware serious cost implications, but there are no simple solutions today. CloudPhysics data shows that it is not uncommon for organizations turn off their HA safety systems because of their complexity. IT teams suffer daily from such operational hazards, which bleed the efficiency of their systems, increasing both cost and risk.

CloudPhysics, the Intelligent IT Operations Management Service
CloudPhysics is uniquely positioned to solve operations management problems because of the collective intelligence it derives from analyzing the world's IT operations data, combined with its unique, patent-pending data center smart modeling and simulation techniques. Customers use CloudPhysics to identify and troubleshoot hundreds of operational issues by accessing focused operational analytics -- referred to as "Cards" -- available on the CloudPhysics SaaS platform. Example capabilities include:
• Health Checks - CloudPhysics acts as an early warning health service for the IT operator to identify operational hazards before they emerge by continuously checking and matching a user's environment against a live database of field issues, including firmware bugs, hardware incompatibilities and other potential problems. CloudPhysics builds the collective intelligence for health checks through its Knowledge Base Advisor, continuously gathering best practices from vendors and field knowledge from the CloudPhysics community of sys admins.
• Reporting - Admins gain insights into their VMware environments through interactive queries using CloudPhysics' Card Builder, which requires no scripting or knowledge of schemas. Users can pull pre-built reports (Cards) shared by community members or build unique ones and share with community members and colleagues.
• Analysis - CloudPhysics gives users big data-driven analytics for storage, compute and other systems to identify radical efficiency and safety improvements for VMware. Customers lower cost and risk by using results based on the CloudPhysics smart modeling techniques.

• Planning - CloudPhysics customers can simulate planned purchases or configuration changes before rolling out in production. For example, CloudPhysics predicts the latency benefit of I/O caching and determines optimal cache sizes to balance cost and performance. Customers run this analysis before making storage hardware purchases (like SSD devices) to avoid wasteful procurement while dramatically improving the performance of business-critical applications. In the case of mission-critical configuration changes common to HA applications, CloudPhysics analyzes HA configuration scenarios before customers commit them into production. Customers optimize resource settings while ensuring critical VMs are protected against server failures.