It is agent less (it gets everything from the HMC/SDMC or IVM).
It supports all kinds of logical partitions: AIX, i5/OS (AS400), Linux, VIOS.
It collects complete physical and logical configuration of all servers and LPAR's.
This program is free software. You can redistribute it and/or modify it under the terms of the GNU General Public License.
For customers under support there is distributed a full version with additional benefits.
Announcements in brief
- Presentation: LPAR2RRD support benefits and business justification
- 2012 summary
- How LPAR2RRD helps in migrations
- 3.30: Active Memory Sharing and enhanced CPU Workload Estimator
- Materials from the 2nd Prague's LPAR2RRD workshop is here:Capacity planning workshop.pdf
- 3.20: Announcement, it brings Custom Groups, Favourites and Alerting.
- 3.15: CPU Workload Estimator, try it on
- 3.14: Live Partition Mobility and generally LPAR migration support has been released.
- 3.10: brings Top10 and new "Physical and Logical configuration" : Summary and Detail examples
- 3.02: comes with yearly trend graphs
- more details on LPAR2RRD announcements
Why use LPAR2RRD?
- 1st of all due to its unique web based GUI for graphical interpretation of physical CPU utilization in virtualized environment. You can on a click find total CPU utilization of any physical box in a simple graphical form understandable from technician to management level
- it is free for use, you do not need to invest into expensive commercial solutions. You can optionally buy support.
- it is agent less == there is no any administration need if there is ANY change in your virtual environment
- it does not require ANY administration, you just put there SDMC/HMC/IVM names, allow ssh access there and that's it!
in comparsion to other commercial tools where you need to have dedicated and trained staff for administration of them
- you can on a click export internally stored data in CVS format for further analysis via 3rd party tools
- for more details check Features section bellow
How it works
- it is abe to graph Active Memory Sharing usage
- CPU Workload Estimator brings possibility to estimate before any LPAR migration if the target HW is cope with CPU load of migrated LPAR(s)
It has option to make estimations and graphs based on rPerf or CPW benchmarks rather than on CPU cores. This allows you to get realistic view when you migrate a LPAR between different IBM Power™ servers.
- You are allowed to group LPARs or CPU pools from different servers into Custom Groups what gives you easy overview of CPU load per the OS clusters, aplication layers etc ...
- It is able to raise an alerts if CPU exceeds defined threshold for a LPAR or CPU pool
- Live Partition Mobility and generally LPAR migration support
- it is able to graph CPU usage prediction 1 year ahead yearly trend graphs
- it is no longer only an off-line tool. You can do refresh (via a WEB link) whenever you want to get nearly "real-time" graphs
- agent-less, no agents need to be installed on LPAR's. It gets all data from HMC/SDMC/IVM servers
- support of all types of LPAR's and OS'es : AIX, VIOS, i5/OS (AS400), Linux on IBM Power™ Systems
- It collects "Physical and Logical configuration" : Summary and Detail examples
- it graphs LAN traffic of all IVE/HEA adapters and SAN traffic (available only for IVM based servers)
- simple to install, configure and use
initial install & configuration together with dependency tools should not take more than an hour!
- no additional management required when you change/add/remove/rename a LPAR or managed system/frame (the tool discovers everything automatically for all already configured HMC/SDMC/IVM)
- data CSV outputs for processing by other tools, include output of server's configuration
- it can be hosted on any *NIX platform, it needs a web server, SSH, Perl and RRDTool installed.
- this is not replacement of tools like IBM Tivoli, Ganglia or Nagios which monitor everything (CPU/MEM/IO/LAN...), it is rather an add-on tool for physical CPU monitoring of IBM Power™ platform which we hope does that particular thing better than the others. Typical usage is for capacity planning purposes, although it is possible to use it also in "real-time" mode.
- it should act as the tool which easy identifies CPU load abnormality and locates it up to the LPAR level. For further deeper analysis is necessary to use AIX OS tools like nmon, topas, trace ...
- it is not a tool for partitions using dedicated CPUs, it always shows 100% CPU utilization for them (unfortunately HMC/IVM/SDMC do not provide detailed CPU stats usage for LPARs with dedicated CPU)
- it does not have a perfect WEB front-end, no nice menus ... functionality and simplicity is primary target.
- IBM DS8k is already under beta testing
- Monitoring of other storages like EMC high ends, IBM Storwize and IBM XIV
It should act as a tool for quick performance check of the storage and for getting trends with intention of capacity planning.
- Graph some basic SAP performance data