EPAM Cloud Orchestration includes a highly informative analytics solution that allows you to monitor the Cloud infrastructure performance and cost alterations at all the structural levels: VM, Project, Region, and Cloud.
This topic contains the following sections:
VM Level metrics allow you to get the performance statistics of your Cloud instances. In EPAM regions, Zabbix and Chef Server info is provided. Due to AWS Cloud Watch integration, you can get the statistics from the VMs located in AWS-* regions. In the current implementation, the VM-level monitoring statistics is not available for instances in Google and Azure clouds.
On Project level, you can monitor the current project costs and their general trends. Here, you can see the influence of different KPI on the overall project cost.
The Region-level analytics allows you to get the performance and cost statistics for the specified EPAM Private Cloud virtualization region. It aggregates the KPI data from all the infrastructure elements existing in this region and provides you with the info on projects and VMs count in the selected region, infrastructure utilization details, and the overall cost of the region.
The Cloud-level analytics allows you to get the performance and cost statistics for the whole EPAM Private Cloud infrastructure. It aggregates the KPI data from all the regions and provides you with the info on projects and VMs count in Cloud, infrastructure utilization details, and the overall Private Cloud cost.
EPAM Cloud Analytics Structure
The provided EPAM Cloud analytics data is represented as a set of graphs and their descriptions, each illustrating the dynamic changes and the general trend of the available Key Performance Indicators (KPI) for the last 7 days. On the picture below, you can see all the four analytics levels and the data they cover:
VM and Project level statistics is available for all EPAM Cloud users and can be reached on Monitoring page of Orchestration UI. The Region and Cloud level statistics is available on the Radar page accessible only for members of the CSUP project.
Analytics Data Representation
Data representation on different levels is not the same. VM-level data representation differs from that of Cloud, Project and Region level and is described in the beginning of VM-Level Analytics section. The details on data representation on other levels are given below.
On each level, the data can be represented in a Summary and a Detailed view. The processed KPI data are grouped by KPI purpose. Each group can be displayed as a summary, giving only the final values, and can be unfolded for the details. From the Detailed view, giving the info on the general KPI value trend for the last 7 days,the user can reach the Deep Dive view of the selected KPI. There, they can find the information on the KPI values in different moments of time, monitor the value changes for up to two months, and compare the values of different KPI.
The values in the Summary view can be displayed in different colors that give the user the idea of the changes in the KPI values:
- Grey: the value has not changed during the last seven days
- Red: the current value is the maximum one for the last seven days
- Yellow: the current value is the minimum one for the last seven days
- Green: the current value is in a "normal" scope, i.e., it does not reach neither maximum, nor minimum point
In Detailed view, each KPI description includes the following information:
- KPI name and description. This includes the KPI name abbreviation used in the Summary view, KPI description and a colored mark indicating the KPI value changes.
- KPI value: On Project level, the KPI Value represents the monthly total value for the selected KPI. On Region and Cloud levels, the Value gives the data for the current day.
- Trend: the graph, representing the KPI trend. In most cases (except for the VM level) the Trend graph does not give the exact numbers. It has an exponential Y-axis to provide clearer visualization in case there is a wide scattering of the values. The main purpose of this graph is to show whether there is a stable growing trend, or a fluctuation around an average value, etc. If the KPI value has not changed during the investigated time period, the trend will be shown as a smooth line.
- Comments: the description of KPI values changes throughout the investigated time period.
VM-Level analytics data representation slightly differs from that of the other levels. Please, see the corresponding section for the details.