Enabling real-time testing of on-demand infrastructure to predict service level agreement compliance
View Patent ↗Provided is an apparatus and method for demonstrating the efficiency and effectiveness of a customer's information technology (IT) system and applications operating in a shared IT, or electronic business on-demand, environment. A test suite mimics hypothetical settings in an on-demand services environment ( 129 ) and then determines how the settings affect the allocation of resources in the on-demand services environment ( 129 ). The calculated allocation of resources is then compared to one or more service level agreements (SLAs) in order to determine compliance with a particular SLA.
1. A computer-implemented method for predicting service level in a utility computing environment having a dynamically allocated subset of computing resources from a set of available computing resources, the method comprising:
creating a resource profile corresponding to the dynamically allocated subset of computing resources allocated according to a service level agreement;
loading a workload profile representing a demand profile for an enterprise;
simulating, by a processor, processing of the workload profile using the resource profile to produce a service level result, wherein the resource profile subset is modified during the simulation according to the service level agreement;
generating a new service level agreement in the event the resource profile cannot process the workload profile at an expected service level corresponding to the service level agreement, wherein the new service level agreement will process the workload profile at the expected service level; and
comparing the workload profile to a second workload profile representing an actual demand profile for a second client account;
wherein the simulating is based upon a result of the comparison.
2. The method of claim 1 , further comprising:
comparing the service level result to the service level agreement; and
signaling whether the resource profile will process the workload profile at an expected service level corresponding to the service level agreement.
3. The method of claim 1 , wherein the subset of computing resources includes allocated processing resources and memory resources for a client account.
4. The method of claim 1 , wherein the service level agreement includes a base resource allocation, a maximum resource allocation, resource costs, and rules for dynamically reallocating the resources based upon workload demand.
5. The method of claim 1 , wherein the simulation is scheduled to run automatically at an off-peak time.
6. The method of claim 1 , further comprising determining a cost associated with meeting the new service level agreement.
7. The method of claim 1 , wherein the resource profile includes a communication bandwidth allocation.
8. The method of claim 1 , wherein the workload profile includes scheduling information and the simulating incorporates the scheduling information in the processing.
9. The method of claim 1 , wherein the workload profile includes information corresponding to one or both of prioritization of resources and importance of specific resources.
10. The method of claim 1 , wherein the workload profile is loaded from a configuration file.
11. A system for simulating service in a utility computing environment having a first service level agreement to service the demands of an enterprise using a dynamically allocated subset of computing resources from a set of available computing resources, comprising:
a processor;
a memory, coupled to the processor;
an allocated subset of the set of computing resources; and
logic, stored on the memory and executed on the processor, for;
loading a workload profile representing a hypothetical demand profile for a client account;
simulating processing of the workload profile, wherein the workload profile is based upon actual, measured data using the allocated subset of the set of available computing resources to produce a service level result;
modifying the allocated subset of the available computing resources based upon the service level result;
generating a new service level agreement in the event the simulation produced by the simulation logic cannot process the workload profile at an expected service level corresponding to the first service level agreement, wherein the new service level agreement will process the workload profile at the expected service level; and
comparing the workload profile to a second workload profile representing an actual demand profile for a second client account;
wherein a simulation produced by the simulation logic is based upon a result of the comparing.
12. The system of claim 11 , the logic further comprising logic for:
comparing the service level result to a second service level agreement; and
signaling the modified, allocated subset of the available computing resources will process the workload profile at an expected service level corresponding to the second service level agreement.
13. The system of claim 11 , wherein the set of available computing resources comprises:
processing resources; and
memory resources.
14. The system of claim 13 , wherein the set of available computing resources further comprises:
a base resource allocation;
a maximum resource allocation;
resource costs; and
rules for dynamically reallocating the resources based upon workload demand.
15. The system of claim 13 , wherein the set of available computing resources further comprises communication bandwidth.
16. The system of claim 11 , wherein the workload profile includes scheduling information and the simulation logic incorporates the scheduling information in the processing.
17. A computer program product for predicting service level compliance in a utility computing environment having a service level agreement to service the demands of an enterprise using a dynamically allocated subset of computing resources from a set of available computing resources, comprising:
a non-transitory memory,
a resource list, stored on the memory for execution on a processor, detailing a set of available computing resources;
an allocated resource list, stored on the memory, detailing an allocated subset of the set of available computing resources; and
logic, stored on the memory for execution on a processor, for:
creating a computer resource profile based upon the allocated subset of the set of available computing resources;
loading a workload profile representing a hypothetical demand profile for a client account;
simulating the processing of the workload profile using the computer resource profile to produce a service level result;
comparing the service level result to a service level agreement;
signaling whether the computing resource profile will process the workload profile at an expected service level corresponding to the service level agreement;
generating a new service agreement in the event the computing resource profile cannot process the workload profile at the expected service level corresponding to the service level agreement, wherein the new service level agreement will process the workload profile at the expected service level; and
comparing the workload profile to a second workload profile representing an actual demand profile for a second client account;
wherein a simulation produced by the simulation logic is based upon a result of the comparing of the workload profile to the second workload profile.
18. The system of claim 17 , wherein the computing resource profile comprises:
processing resources; and
memory resources.
19. The system of claim 18 , wherein the computing resource profile further comprises:
a base resource allocation;
a maximum resource allocation;
resource costs; and
rules fur dynamically reallocating the resources based upon workload demand.
20. The system of claim 18 , wherein the computing resource profile also comprises a communication bandwidth allocation.
21. The system of claim 17 , wherein the workload profile includes scheduling information and the simulation logic incorporates the scheduling information in the processing.