Community analysis of computing performance
Embodiments of systems and methods are described for performing community analysis of computing performance. These systems and methods can collect and analyze community performance data from a plurality of physical or virtual infrastructure installations. The systems and methods can leverage this community performance data to identify and recommend areas where individual physical or virtual infrastructures can improve performance.
1. A method of providing community analytics related to a plurality of tenant systems, the method comprising:
collecting data from a plurality of tenant systems, at least some of the tenant systems being operated by different entities, the plurality of tenant systems constituting a community of tenant systems associated for the purpose of sharing comparative analytics about the tenant systems;
computing analytics from the data collected from the tenant systems, the analytics comprising statistics that reflect shared aspects of the tenant systems;
receiving a request for analytics from a user of a first one of the tenant systems;
in response to receiving the request, outputting for presentation to the user both a portion of the collected data regarding the first tenant system together with at least a portion of the computed analytics for comparison, thereby enabling the user to compare aspects of the first tenant system with the analytics associated with the community;
wherein at least said computing the analytics is implemented by a computer system comprising computer hardware;
identifying top performers of the tenant systems, other than the first tenant system, on one of the computed analytics;
comparing differences between characteristics of the top performers and the first tenant system;
identifying one or more characteristics for which a statistically-significant number of the top performers have a similar configuration; and
outputting, to the user of the first tenant system, a representation of the one or more characteristics as recommendations for potential configuration change to the first tenant system.
2. The method of claim 1 , wherein the shared aspects of the tenant systems comprise one or more of the following: virtual machine information, host information, workload, central processing unit (CPU) usage, and memory usage.
3. The method of claim 2 , wherein the analytics comprise one or more of the following: an average number of virtual machines in at least some of the tenant systems, an average number of virtual machines per host, average CPU usage per virtual machine, average memory usage per virtual machine, used CPU amount for a top CPU consumer, consumed memory of a top memory consumer, and storage capacity.
4. The method of claim 1 , wherein said collecting, said receiving, and said outputting are implemented by the computer system.
5. The method of claim 1 , wherein the computer system comprises a plurality of physical computing devices.
6. A system for providing community analytics related to a plurality of tenant systems, the system comprising:
a data collection module configured to collect data from a plurality of tenant systems;
a community analytics module implemented by one or more processors, the community analytics module configured to compute analytics from the data collected from the tenant systems with one or more processors, the analytics comprising statistics that reflect shared aspects of the tenant systems; and
a user interface module configured to output for presentation to the user both a portion of the collected data regarding the first tenant system together with at least a portion of the computed analytics for comparison, thereby enabling the user to compare aspects of the first tenant system with the analytics associated with the community;
a comparative analysis engine configured to generate performance recommendations that recommend configuration changes to the tenant system related to the computed analytics; and
wherein the comparative analysis engine is further configured to generate the performance recommendations by comparing differences between configurations of second ones of the tenant systems that perform better than the first tenant system on a first of the computed analytics.
7. The system of claim 6 , wherein the comparative analysis engine is further configured to generate the performance recommendations by ranking the configurations of the second tenant systems based at least in part on a degree of confidence that the configurations cause better performance on the first computed analytic.
8. The system of claim 6 , wherein the comparative analysis engine is further configured to generate the performance recommendations by ranking the configurations of the second tenant systems based at least in part on categories of the configurations and a category of the first computed analytic.
9. Non-transitory physical computer storage comprising instructions stored thereon that, when executed in one or more processors, implement operations for providing community analytics related to a plurality of tenant systems, the operations comprising:
collecting data from a plurality of tenant systems, the tenant systems being operated by different entities, the plurality of tenant systems constituting a community of tenant systems associated for the purpose of sharing comparative analytics about the tenant systems;
computing analytics from the data collected from the tenant systems with one or more processors, the analytics comprising metrics that reflect shared aspects of the tenant systems; and
outputting for presentation to a user of a first one of the tenant systems both a portion of the collected data regarding the first tenant system together with at least a portion of the computed analytics for comparison, thereby enabling the user to compare aspects of the first tenant system with the analytics associated with the community; and
generating the performance recommendations that recommend configuration changes to the tenant system related to the computed analytics;
wherein the performance recommendations are generated by comparing differences between configurations of second ones of the tenant systems, other than the first tenant system, that perform better than the first tenant system on a first of the computed analytics.