CloudPhysics updates, now with more awesome included
On the first day of VMworld 2014 in San Francisco, CloudPhysics kicks the conference off with a big announcement. The patform is receiving another round of new features that will blow your socks off. I’ve followed CloudPhysics closely over the past couple of years since they announced themselves to the world. Given the way they built their hosted offering I immediately began to dream of what might be possible. Some of the new features are exactly what I dreamed of and others are just more awesomeness that they dreamed up.
Global Data Set
CloudPhysics has been collecting data from every environment that installs their observer appliance. The observer is necessary to use the CloudPhysics offering, this is how data is collected and uploaded. The tools then crunch the data, for a lack of better description. You are able to use your data for reports and monitoring. CloudPhsyics also has been collecting two years worth of annonomized customer data that is now like the hive mind.
No one else has this level of information about virtualization. Oh the things that could be done with this, well I’m glad to say that they now have enough unique data that represents unique environments of all sizes that can be used for comparison purposes. I can’t tell you how cool this is going to be.
With this global data set CloudPhysics can now offer Global Insights that can compare values in your environment to the averages. This part makes my mind go crazy with the possibilities. Think of the easy ones like compare my cpu ratios, vm consolidation, memory over commit are a few that come into mind. I’m still working on finalizing what values will be immediately available and there are talks about what else would be of value that could be offered with this type of data. The image below shows a sample of what is possible.
Initially this feature is only be a technology preview, but they wanted to show what they are working on. I think of this as bringing the power of vscsiStats to the masses. Irfan Ahmad the guy that created vscsiStats and several of the advanced storage functions in vSphere has created this feature for CloudPyshics. As CTO of CloudPhysics Irfan is using his experience to build technology that can help architects and admins diagnose and design for complex storage problems.
The following is a short description of the new feature. It is taking in the data and creating visual representations of the data so that you can easily diagnose workload characteristics. This will help in troubleshooting storage issues or when designing for a new storage environment.
CloudPhysics analyzes and characterizes storage workload spatial locality (sequentiality vs randomness), dominant I/O block sizes, and the complete latency profile, then visualizes these into “workload shapes.” Administrators quickly learn which shapes are “normal” for their environment and identify outliers that indicate performance troublespots, accelerating time to resolution. This deep visibility comes from unique underlying technology, in which every I/O is analyzed, ensuring anomalies don’t get lost in the averages, which is what happens with most other tools.
Improved Alerting and Reporting
Along with the other new features CloudPhysics continues to improve their reporting abilities and their performance monitoring capabilities. Below are short summaries of the other features being updated in the new release.
Enhanced Performance Troubleshooting: Building on the recently released Datastore Contention analytic, which provides insights into disk I/O contention at the datastore level, CloudPhysics’ new VM Disk I/O Contention analytic focuses on VM-level performance. An interactive timeline visually correlates patterns among datastores/VMs, and dramatically simplifies the exploration of hotspots. The algorithm used to detect contention has been tuned and validated using CloudPhysics’ Collective Intelligence.
Smart Alerts: “Guest Partition in VM Running Out of Space” and “Unused VMs” further expand CloudPhysics’ set of SmartAlerts, first introduced in June. CloudPhysics evaluates all objects in the virtual datacenter against certain criteria (e.g., latency, duration, outstanding IOs, IOPS, etc.) and triggers Smart Alerts based on thresholds derived dynamically from patterns and trends observed across our global dataset. Users foresee when conditions are degrading and receive specific recommendations for preemptive measures.
About Brian Suhr
Brian is a VCDX5-DCV and a Sr. Tech Marketing Engineer at Nutanix and owner of this website. He is active in the VMware community and helps lead the Chicago VMUG group. Specializing in VDI and Cloud project designs. Awarded VMware vExpert status 6 years for 2016 - 2011. VCP3, VCP5, VCP5-Iaas, VCP-Cloud, VCAP-DTD, VCAP5-DCD, VCAP5-DCA, VCA-DT, VCP5-DT, Cisco UCS Design