The current row over which of IBM and HP boasts the most energy efficient Blade systems could be heading for a resolution after both vendors said they would welcome tests overseen by an independent and credible third-party.
The row started earlier this month when IBM issued claims that internal testing had shown its AMD Opteron-based BladeCenter (LS21) used within the BladeCenter system requires 30 percent less energy when idle than the comparable HP Opteron-based BladeSystem (BL465c) in the HP cClass system. The company added that its server uses 18 percent less energy when running at full load. Meanwhile, IBM's Xeon-based BladeCenter systems also enjoys similar savings compared to HP models, according to its tests.
HP responded in a statement claiming that: "Based on what we know about IBM’s aging architecture, we feel strongly that these numbers are not a true apples-to-apples comparison".
The company went on to argue that its systems have more standard features and that "if IBM were able to match HP blades in this regard, we would be happy to have the HP BladeSystem tested against their best effort by an independent, unbiased and credible third-party".
However, Wayne Flaggs, vice president of enterprise BladeCenter at IBM, dismissed suggestions that the tests had been flawed. "It was pretty simple really," he said. "We put the same configurations side by side, ran them and metred the electricity."
He also insisted IBM was happy to take up HP's challenge of independent tests. "If that is what they're [HP] saying then bring it on," he said. "We'd fully support third party testing."
Flaggs accepted that the difficulty in staging such tests - as with any server bake off - would revolve around picking agreed server configurations. But he added that this was not an insurmountable problem and that it should be possible for a third party to put together a number of different configurations capable of mirroring real world scenarios.
"Energy efficiency is the biggest issue IT directors face and the more data there is available the better," he said. "The smart ones will understand what configurations were used and which best suit their scenario."
He added that if agreeing on a third party to undertake the tests proved tricky then using several different bodies could help convince customers that the tests were fully independent.
Of course both companies can duck out of this potential joust by arguing that the different standard features on each system mean a true "apples for apples" test is impossible. But given that no two products from rival vendors will ever be identical this strikes as a pretty thin argument, particularly when thorough testing should be able to find a way to account for any differences.
The onus is now on both vendors to back up their talk of independent tests, pick up the phone and set a date for the big fight.
Survey on eco-concerns shows older people more worried about plastic waste, while younger people cite global warming as top green challenge
New MIT study suggests cost of climate policies targeting CO2 reduction will be more than cancelled out by reductions in healthcare costs from air pollution
Preliminary results from Starbucks trial suggests charging customers extra for takeaway cups is an effective way to change behaviour
Robertsbridge's Charles Secrett takes a tour d'horizon across the UK's green economy, and asks if mainstream businesses are ready for the disruption ahead