A few years ago, I attended an Intel Labs presentation and one of the more interesting segments was on a technology it was quietly developing for large network switches. Intel argued that it could do to the very expensive and high-margin switch business what it did to UNIX servers over the last two decades to cut costs dramatically.
Apparently, Intel has now started executing on that strategy with the acquisition of Fulcrum Microsystems, a fabless semiconductor and related software vendor, targeting low-cost, high-performance, high-end switches.
This will be good news for HP, but bad news for Cisco. Let me explain.
Cisco: The Sun of Switch Vendors
Sun was the big loser in the move to x86 and lower-cost servers. It was also the big beneficiary of large-scale UNIX servers that cut into the mainframe and mid-range computer market like a hot knife through butter. However, ironically, it was IBM, HP and newcomer Dell that benefited from the move to x86, and Sun first tried to fight the move and then, after it was too late, tried to embrace it, eventually failing and being bought by Oracle.
At the core of this problem was the reality that Sun didn't want to sacrifice its high-margin products for low-margin competitors and failed to recognize, until too late, that the company didn't really get a vote. The market was moving with or without Sun, which, ironically, was the same difficult decision Sun had presented to IBM that contributed to IBM's stumble in the late 1980s. (As a side comment, it is really interesting how the technology market repeats cycles and how many technology companies treat each similar cycle as if it were a surprisingly new and unique event. Kind of like being surprised when the sun comes up every morning.)
So the cycle we are talking about here is one where a chip vendor, traditionally Intel (though it is happening to some extent to Intel by ARM in smart devices), comes in and finds a way to commoditize a high-margin business. The dominant provider stumbles because it can't adjust to the new lower-margin reality.
Clearly that would hurt Cisco, but why would that help HP?
HP may not be alone here, as I expect Dell to get on this bandwagon relatively quickly as well. HP is a challenging vendor having bought 3COM as a response to Cisco's entering the server business and HP has been cutting into Cisco's market share and profits significantly. In fact, the coming 6,500 layoffs that Cisco has announced appears to be collateral damage from Cisco's decision to change HP from a partner to a competitor. (Oracle should pay attention as these outcomes tend to come several years after the trigger event.)
HP is already on a commoditization path and is using that as its primary weapon against Cisco. HP is already a close partner with Intel on x86 and Itanium so it wouldn't be hard for them to embrace this new Fulcrum technology quickly and even more aggressively cut prices in order to cut even more deeply into Cisco's market share and profits. Dell is on a similar total provider path and it seems likely it will see the blood in the water as well and move on the opportunity if only as a competitive hedge against HP. But Cisco would be the vastly easier and more attractive target given its relative size in this market.
The end result, if Cisco didn't aggressively move on the same opportunity to avoid Sun's mistake, would be a major change in market makeup and power. Though it would be incredibly ironic if Oracle (assuming it is still around) ended up buying Cisco at the end of this cycle as it did Sun, though I think that is unlikely.
When Intel enters a market, as Sun learned against a vastly weaker Intel, it tends to change the market dynamic. Long-time Intel partners like HP and Dell should benefit from this move and the current dominant player, Cisco, will be put at risk by it. The question now is whether Cisco has learned from Sun's failure and can avoid it or whether it will repeat the same mistakes. As noted above, the interesting thing is that executives don't seem to want to learn by watching others and seem to enjoy making these mistakes themselves. That doesn't bode well for Cisco, but would be a nice present for Intel, HP and Dell assuming I've read this right.