IBM System x3755 M3 review
IBM’s latest System x3755 M3 is its first rack server to support AMD’s Opteron 6100 processors. It’s taken a very long time to materialise and in this exclusive review we tell you whether it’s a better bet than the four-socket AMD servers from HP and Dell.
IBM scored well in our power tests. With Windows Server 2008 R2 in idle we measured the x3755 drawing a very modest 214W. Using SiSoft Sandra to push all 48 physical cores to near maximum utilisation saw the power draw peak at 552W.
The x3755 uses noticeably less power than its closest competitors. A Dell R815 with quad 2.2GHz Opteron 6174 processors and 64GB of memory was measured at 335W in idle and 618W under load. An HP DL585 G7 with four 1200W supplies and the same processors, but only half the memory, recorded 356W in idle and a much higher peak of 710W.
Build quality of the x3755 is extremely good and easily a match for Dell and HP. Internal cooling is handled by a bank of five hot-swap fans and these are accessed via a small hatch in the server's lid. Due to the internal layout you can't run the server with the whole lid removed and doing so will initiate an immediate system shutdown.
Internal design is quite unusual and had been largely dictated by the power supply canister which sits on top of the motherboard. This is hinged at the front and can be easily released and swung up to reveal the processor and memory sockets underneath.
An expansion cage is located to the left side and IBM's tool-less design makes it very easy to remove. The M1015 RAID card is fitted in a dedicated slot in the cage's riser and there are three more PCI-E slots at the back with all accepting half-height, half-width cards.
The 3D skills report
Add 3D skills to your creative toolkits and play a sizeable role in the digital future

The increasing need for environmental intelligence solutions
How sustainability has become a major business priority and is continuing to grow in importance

2022 State of the multi-cloud report
What are the biggest multi-cloud motivations for decision-makers, and what are the leading challenges

Solve global challenges with machine learning
Tackling our word's hardest problems with ML
