BonFIRE logo and link to main BonFIRE site

Table Of Contents

Previous topic

BonFIRE OVF Schema Definition

Next topic

Release Notes

This Page

Default Monitoring Metrics

VM Metrics

The following metrics are actively measured by default:

Buffers memory
Cached memory
CPU system time (avg1)
CPU nice time (avg1)
CPU idle time (avg1)
CPU iowait time (avg1)
CPU user time (avg1)
Free disk space on /
Free memory
Free swap space
Host boot time
Host status
Host uptime (in sec)
Incoming traffic on interface lo
Incoming traffic on interface eth1
Incoming traffic on interface eth0
Number of processes
Number of running processes
Number of users connected
Outgoing traffic on interface lo
Outgoing traffic on interface eth0
Outgoing traffic on interface eth1
Ping to the server (TCP)
Processor load
Shared memory
System cpu usage average
Total disk space on /
Total memory
Total swap space
Used disk space on /
Used disk space on / in %

The following metrics are disabled (are not measured by default), but can be enabled by BonFIRE user any time:

Checksum of /usr/sbin/sshd
Checksum of /usr/bin/ssh
Checksum of /vmlinuz
Checksum of /etc/services
Checksum of /etc/inetd.conf
Checksum of /etc/passwd
Email (SMTP) server is running
Free disk space on /usr
Free disk space on /var
Free disk space on /tmp
Free disk space on /home
Free disk space on /opt
Free disk space on /tmp in %
Free disk space on /var in %
Free disk space on /usr in %
Free disk space on / in %
Free disk space on /home in %
Free disk space on /opt in %
Free number of inodes on /usr
Free number of inodes on /tmp
Free number of inodes on /home
Free number of inodes on /
Free number of inodes on /opt
Free number of inodes on /tmp in %
Free number of inodes on / in %
Free number of inodes on /usr in %
Free number of inodes on /opt in %
Free number of inodes on /home in %
Free swap space in %
FTP server is running
Host information
Host local time
Host name
IMAP server is running
Maximum number of opened files
Maximum number of processes
News (NNTP) server is running
Number of running processes zabbix_server
Number of running processes zabbix_agentd
Number of running processes apache
Number of running processes inetd
Number of running processes mysqld
Number of running processes sshd
Number of running processes syslogd
POP3 server is running
Processor load5
Processor load15
Size of /var/log/syslog
SSH server is running
Temperature of CPU 1of2
Temperature of CPU 2of2
Temperature of mainboard
Total disk space on /home
Total disk space on /usr
Total disk space on /tmp
Total disk space on /opt
Total number of inodes on /usr
Total number of inodes on /
Total number of inodes on /opt
Total number of inodes on /home
Total number of inodes on /tmp
Used disk space on /usr
Used disk space on /var
Used disk space on /home
Used disk space on /tmp
Used disk space on /opt
Used disk space on /usr in %
Used disk space on /var in %
Used disk space on /tmp in %
Used disk space on /opt in %
Version of zabbix_agent(d) running
WEB (HTTP) server is running

Infrastructure Metrics

Information about a number of predefined metrics that are provided to BonFIRE experimenter about the physical machines hosting their VMs. BonFIRE solution allows each testbed to dynamically provide their own templates. Those of common interest are for example as follow:

Eth0 outgoing traffic
Eth0 incoming traffic
Running VMs
Processor load
Free swap space
Total memory
Free memory
Disk sda Write Bytes/sec
Disk sda Write: Ops/second
Disk sda IO ms time spent performing IO
Disk sda IO currently executing
Disk sda Read: Milliseconds spent reading
Disk sda Read: Ops/second
Disk sda Write: Milliseconds spent writing
Disk sda Read Bytes/sec
Ping to the server (TCP)

ECO Metrics

Through their involvement in the ECO2Clouds project, the Inria, EPCC, and HLRS sites provide further monitoring information about energy usage and CO2 estimation. The related metrics are as follows:

Energy Mix

Metric Definition Unit
Biomass, CCGT (Combined Cycle Gas Turbine), Coal, Cogeneration (of heat and power), Fossil, Gas, Geothermal, Hydraulic, NPS hydro, Nuclear, OCGT (Open Cycle Gas Turbine), Oil, Other, Pumped storage, Solar, Total green, Water and Wind Energy sources %
Grid Total (only available at Inria and EPCC) Total power generated for the national electricity Grid MW
Imported, exported How much electricity is imported/exported %
CO2 per kWh How much CO2 is emitted per kWh generated g/kWh

Site

Metric Definition Unit
Site utilization Current utilisation of a single site. Defined as (available cores) / (total cores). %
Storage utilization Percentage of the frontend storage used %
Availability If the OCCI server provides a reasonable answer to a request AND at least one host has one available core) Boolean
PUE Power Usage Effectiveness: the ratio between the total facility power and the power that is used by the computing equipment. None

Host

Metric Definition Unit
Power consumption The power consumed by the analysed host in a specific time period. W
Disk IOPS the I/O operations of the disk within a host. IOPS/s
CPU utilization The average utilization of the processors inside a host %
Availability   Boolean