== Provisioning of resources (3pts) == 3 Experimental methodology and results show evidence of student's ability to provision both compute and storage resources in EC2 and compute resources in Azure 2 Experimental methodology and results show evidence of student's ability to provision two of the three types of resources (compute in EC2, storage in EC2, compute in Azure) 1 Experimental methodology and results show evidence of student's ability to provision one of the three types of resources (compute in EC2, storage in EC2, compute in Azure) 0 No resources were provisioned == Experiment design (4pts) == 4 Experimental methodology illustrates the capabilities and limitations of compute, network, and storage resources in isolation, accounting for confounding factors and interactions between the resources 3 Experimental methodology illustrates the capabilities and limitations of compute, network, and storage resources, but does not account for confounding factors and interactions between the resources 2 Experimental methodology only illustrates the capabilities and limitations of two of the three types of resources (compute, network, and storage) 1 Experimental methodology only illustrates the capabilities and limitations of one of the three types of resources (compute, network, and storage) 0 No experimental methodology is provided == Experiment description (3pts) == 3 Experimental methodology is sufficiently detailed, indicating what tools, EC2/Azure resources, and parameter values were used, such that a user familiar with these cloud services could reproduce the experiments 2 Experimental methodology lacks some details, making it challenging for a user familiar with EC2/Azure to reproduce the experiments 1 Experimental methodology lacks significant details, making it impossible to reproduce the experiments 0 No experimental methodology is provided == Compute performance measurements (4pts) == 4 All of the following 4 questions are answered with insightful, accurate explainations supported by measurement results: What is the compute performance of specific types of EC2 instances? Is performance increase commensurate with resource increase? How does EC2 compute performance compare to your local machine? How does the compute performance in EC2 and Azure compare? OR 3 of the above questions plus 1 alternative, interesting compute performance question are answered with insightful, accurate explainations supported by measurement results 3 3 of the above questions (or 2 of the above questions plus 1 alternative question) are answered with insightful, accurate explainations supported by measurement results 2 2 of the above questions (or 1 of the above questions plus 1 alternative question) are answered with insightful, accurate explainations supported by measurement results 1 1 of the above questions (or 1 alternative question) is answered with insightful, accurate explainations supported by measurement results 0 The measurement results do include anything about compute performance == Network performance measurements (4pts) == 4 All of the following 4 questions are answered with insightful, accurate explainations supported by measurement results: What is the network performance between the same type of instances? What is the network performance between different types of instances? Is network performance consistent over time? How does the network performance in EC2 and Azure compare? OR 3 of the above questions plus 1 alternative, interesting network performance question are answered with insightful, accurate explainations supported by measurement results 3 3 of the above questions (or 2 of the above questions plus 1 alternative question) are answered with insightful, accurate explainations supported by measurement results 2 2 of the above questions (or 1 of the above questions plus 1 alternative question) are answered with insightful, accurate explainations supported by measurement results 1 1 of the above questions (or 1 alternative question) is answered with insightful, accurate explainations supported by measurement results 0 The measurement results do include anything about network performance == Storage performance measurements (4pts) == 4 All of the following 4 questions are answered with insightful, accurate explainations supported by measurement results: How much performance variation is there in regular EBS over time? How much performance variation is there in regular EBS over space? Does the performance of provisioned IOPS match advertised values? Do the capabilities of a compute instance affect storage performance? OR 3 of the above questions plus 1 alternative, interesting storage performance question are answered with insightful, accurate explainations supported by measurement results 3 3 of the above questions (or 2 of the above questions plus 1 alternative question) are answered with insightful, accurate explainations supported by measurement results 2 2 of the above questions (or 1 of the above questions plus 1 alternative question) are answered with insightful, accurate explainations supported by measurement results 1 1 of the above questions (or 1 alternative question) is answered with insightful, accurate explainations supported by measurement results 0 The measurement results do include anything about storage performance == Measurement results (3pts) == 3 Measurement results are presented in a clear format and explained in the text 2 Measurement results either are not presented in a clear format or are not explained in the text 1 Measurement results are hard to understand 0 No measurement results are presented