This page lists the views available in the explore page. Use them to change the context of the explore page and focus on a particular scope such as CPU, Requests, Network and many others. Play with them using the left sidebar in the explore page.
Overview of general resource consumption metrics
This view offers a general overview of the health of your infrastructure, with common resource utilization metrics like cpu, memory, file, and network utilization all in a single pane of glass.
Usage tip: Try grouping by host and/or container metadata when using this view (eg: host.mac > container.id). This view is often most useful when exploring groups of hosts and containers, since the metrics are relevant to underlying resource consumption.
CPU/Memory (Container Limits)
CPU and memory utilization metrics for containers
This view breaks down both CPU and memory metrics for containers in two distinct ways:1. Resource utilization of containers relative to the total host resources.2. Resource utilization of containers relative to the host resources allocated to each container.
Usage tip: CPU shares and quotas represent two different ways that CPU can be allocated to a container at the kernel level. For detailed descriptions of these metrics, see the <a href='http://support.sysdigcloud.com/hc/en-us/articles/204931155#container' target='_blank'>Sysdig metrics dictionary.</a>
CPU and memory usage metrics
This view offers detailed metrics on CPU and memory usage, including stolen CPU time, memory faults and paging.
Usage tip: This view can help you determine if your instances are properly sized for the application in terms of CPU and memory capacity. High 'Memory Usage %' along with excessive hard page faults and large amounts of swapping can be good indicators that more memory or application performance tuning is required.
Breakdown of transaction times
This view offers a breakdown of where application transactions are spending the most time. The time transactions are spending in the current tier is broken out into processing time, file i/o time, and network time. Then a separate calculation shows the percentage of the transactions’ total time that was spent in following tiers.
Overview of network-level metrics
This view offers a detailed view into your network health, with common network-level metrics like network I/O and connection counts all in a single pane of glass.
Overview of file system metrics
This view offers a detailed view into your file system health, with common file-level metrics like file I/O and file system usage all in a single pane of glass.
Overview of service-level metrics
This view offers a general overview of the health of your applications, with common service-level metrics like container count and request activity all in a single pane of glass.
Usage tip: Try grouping by application and service metadata when using this view (eg: container.image > container.id). This view is often most useful when exploring groups of applications and services, since the metrics are relevant to holistic service performance.
Estimated performance capacity.
This view shows CPU% capacity usage and transaction requests and estimates the loss of capacity to service requests due to 'stolen cpu'. Estimated Max Requests are for the current CPU capacity, while Estimated Stolen Requests are those additional requests the instance could handle were it able to utilize the full physical CPU. Metrics:
- Capacity in Use %
- Capacity Stolen
- Reqs In
- Estimated Max Reqs
- Estimated Stolen Reqs
Usage: By default, instances are shown with processes in aggregate, be sure to click the '+' symbol by each instance to see capacity metrics for individual processes.
Tip: CPU steal time' is a measure of the percent of time that a virtual machine's CPU is in a state of involuntary wait due to the fact that the physical CPU is either shared among virtual machines or due to contracted limits.