Installation guide

20 Operating Manual RX300 S7
Functional overview
Cooling
The Cool-safe™ cooling concept with improved air flow cooling technology
(honeycomb design) ensures the highest possible performance of the
processors at work; at the same time, the system is extremely reliable thanks to
the reduced heat dissipation.
The server is equipped with five separate hot-plug fan modules. The redundant
fan configuration allows one fan to fail and the system still works. To ensure the
redundancy of the fan system the faulty fan has to be replaced immediately.
Defective fans can be replaced during operation (for further details see the
"PRIMERGY RX300 S7 Server Upgrade and Maintenance Manual").
Each processor has a heat sink.
High level of availability and data security
When memory data is accessed, 1-bit errors are identified in the main memory
and automatically corrected with the error correcting code (ECC) method. The
patented memory scrubbing function regularly starts up the EDC mechanism,
ensuring continuous data integrity.
The memory modules used support SDDC technology (Chipkill™), which
further increases the effectiveness of memory error monitoring and correction.
Rank sparing memory technology is also supported, whereby one memory
module is used as a replacement for two other memory modules. If a memory
module fails, the module with the defective memory is deactivated automatically
and the replacement module is activated (provided that it has been configured
accordingly in BIOS; see BIOS Setup). The deactivated memory module is no
longer used and the faulty memory module can be replaced at the next
opportunity. The spare rank module must have identical or larger memory
capacity than all the other ranks on the same channel.
Memory modules can be protected against failure by mirroring (Memory
Mirroring). Memory Mirroring is roughly comparable to RAID level 1 for hard
disk arrays.
I For further information according to memory module population please
refer to the technical manual of the system board D2939.
ASR&R (Automatic Server Reconfiguration and Restart) restarts the system in
the event of an error and automatically "hides" the defective system
components.