Computer system performance data recorder
Kronometrix Data Recording, is an automatic, simple to use, data collection package. It has a small memory and system footprint, available for ICT enterprise and industrial IoT.sysrec, the main ICT data recorder, reports computer system performance, including the most important metrics from several operating systems, like: RedHat, CentOS, Debian, FreeBSD, Windows or MacOS. To be able to record data as accurate as possible, the recorder should run on each host or operating system, we plan to collect data from.For example, under Linux based operating systems, the recorder, uses different system interfaces, like PROCFSor SYSFS to fetch important performance metrics, like: overall CPU utilization, across all virtual or physical CPUs, including SYSTEM, USER and IDLE time, the system run queue length, memory statistics, disk and network throughput and errors.
FIELDS:
CPU
#01 timestamp : seconds since Epoch, time
#02 cpupct : cpu utilization, across all cpus, number
#03 sumpct : sum of all cpus utilization, number
#04 headpct : headroom cpu available, all cpus, number
#05 userpct : cpu utilization, user space in percent, number
#06 nicepct : cpu utilization, user space with nice priority, number
#07 sysct : cpu utilization, system space, number
#08 idlepct : cpu utilization, idle state, number
#09 iowaitcpt : cpu percentage in idle state because an I/O operation
is waiting to complete, number
#10 irqpct : cpu percentage servicing interrupts, number
#11 softirqpct : cpu percentage servicing softirqs, number
#12 stealpct : cpu percentage of time spent in other operating systems
when running in a virtualized environment, number
#13 runqsz : run queue length, tasks waiting for run time, number
#14 plistsz : tasks in the task list, number
MEM
#15 memusedpct : size of used memory in percent, number
#16 memused : size of used memory in kilobytes, number
#17 memfree : size of free memory in kilobytes, number
#18 memtotal : size of memory in kilobytes, number
#19 buffers : size of buffers used from memory in kilobytes, number
#20 cached : size of cached memory in kilobytes, number
#21 realfree : size of memory is real free, number
(memfree+buffers+cached)
#22 realfreepct: size of memory real free in percent of total memory, number
#23 swapusedpct: size of used swap space in percent, number
#24 swapused : size of swap space is used is kilobytes, number
#25 swapfree : size of swap space is free in kilobytes, number
#26 swaptotal : size of swap space in kilobytes, number
#27 swapcached : memory that once was swapped out, is swapped back in
but still also is in the swapfile, number
DISK
#28 reads : disk read requests per second, rate
#29 rkbytes : read KB per second, rate
#30 writes : disk write requests per second, rate
#31 wkbytes : write KB per second, rate
#32 iops : disk read+write requests per second, rate
#33 rwkbytes : read+write KB per second, rate
NIC
# rx received, inbound
#34 rxpkt : rx packets per sec, rate
#35 rxkbytes : rx KB per sec, rate
#36 rxerr : rx packets containing errors, rate
#37 rxdrop : number of rx packets that were dropped per second, rate
# tx transmitted, outbound
#38 txpkt : tx packets per sec, rate
#39 txkbytes : tx KB per sec, rate
#40 txerr : tx packets containing errors, rate
#41 txdrop : number of tx packets that were dropped per second, rate
# throughput
#42 rxtxpkt : rx+tx packets per sec, rate
#43 rxtxkbytes : rx+tx KB per sec, rate
#44 avg1 : LA of the last minute, number
#45 avg5 : LA of the last 5 minutes, number
#46 avg15 : LA of the last 15 minutes, number
sysrec – all current reported metrics, grouped by functional categories, CPU, Memory, Storage and Network
Kronometrix vs. SNMP monitoring
We did not make Kronometrix, yet another SNMP network monitoring software. We wanted to build something new, applicable to different industries, which can bring immediate value by seeing and understanding data and the workloads. To achieve that, we have designed and developed a different data recording and analysis pipeline, right from start:
- we have carefully selected and analyzed, for each supported industry, the most needed metrics for different business cases
- then, we have grouped and categorized these metrics by their usage and importance, filtering out non essential parameters
We have find out, that we don’t need to record thousands of metrics and parameters to run a data analysis process. In fact, we would need very few metrics, very important to store and process. Classifying and groping these metrics, was more than enough to have a very efficient and intelligent data analysis pipeline.
On the other hand, traditional SNMP monitoring applications, offer a long list of metrics, probes or sensors which can be activated and displayed, without any deep understanding of their importance, relation and classification. Such systems, end-up just a collection of widgets, charts or other visual information hard to be used for data consolidation and analysis.
For example, lets take a look how other SNMP network monitoring system, Paessler, displays computer system usage, in terms of CPU utilization, for example. You can easily see a list of CPU performance metrics, like utilization, but without any possibilities to drill-down and find more detailed information, like CPU SYSTEM or USER time, or how the CPU usage relates to the memory or storage consumption.
Visualize Kronometrix data
Kronometrix, using data recorders, will help to collect the most important metrics from your systems and applications, ready for data analysis and decision-making. For example, this is how Kronometrix reports sysrec data, using two simple built-in applications: Ops and System reporting:
- overall system performance in terms of utilization and throughput
- CPU, memory, storage and network resources
- run-queue length
- total system capacity processes
We will soon publish a new video blog about our data recorders, where you will learn about the main recorders and how to use them. If you want to know more about our sysrec, or Kronometrix Data Recording please contact us or join the discussion.