@selinz: It's unfortunate to hear that the company that was arguably the greatest developer of electronic hardware will exist the hardware business completely.
Well, they haven't yet, and I don't expect that to happen for some time.
But you can look at it as another facet of the forces that made various vendors go "fabless". As hardware becomes a commodity, it becomes increasingly difficult to make money selling it, and as the costs of building the plant to make the hardware skyrocket, it becomes increasingly difficult to justify the capital expenditure.
Even someone like IBM will say "We can't make money doing this, so why continue to do it? Sell the business to someone else."
IBM will increasingly develop designs someone else will actually build, and make thier money licensing the IP.
I don't really think that's sad - just the natural life cycle of the business.
@LarryM99: They are building their own servers, cutting the 'established' vendors out of the loop.
I think it's fairer to say they are designing their own servers. Neither Facebook nor Google is really in the hardware business (though Google is dipping its toes into those waters.)
If Facebook decides it needs another data center with 10,000 servers to meet demand, they'll contract with an OEM partner to supply them. So will Google. But both will have specific ideas of what they want the servers to be, and will present a set of specs and designs the OEM will be expected to work from.
The question is who the OEM partner(s) will be, and the answer will probably be "Whoever offers the best price, with a track record that assures us they can build systems at that price that will meet the specs for reliability."
@DMCunney, it is very appropriate that you mention Facebook and Google, and you can add other cloud vendors to the list. They are building their own servers, cutting the 'established' vendors out of the loop. Given this, what IBM is selling is definitely a commoditized business with only the customers who aren't virtualized ... yet.
@rick merritt: I think Foxconn and Quanta have the inside track on the big data center server busienss these days.
Possible. I don't know. For example, I think of Foxconn as doing assembly of consumer products like iPhones, and don't have a good feel for what other pies they may have fingers in,
If I'm Facebook or Google, I'll give preference to people who are established server manufacturers, like Dell or HP. Servers have higher reliability requirements than consumer products, because server failures cost more to those who run them. Since both of those have manufacturing done in China to get lower costs, the question becomes who does it for them? Do they have wholly owned facilites, or are they contracting out in an OEM deal? (I bet on the latter.)
If Lenovo can offer the pricing desired, selling a line of already established servers, I think they have a shot.
@DMcCunney: I think Foxconn and Quanta have the inside track on the big data center server busienss these days. Anyone know how well set up Lenovo is with its own board and system manufacturing? Are they winning these big data center deals?
There is a big opportunity there to serve the market of big cloud and data center providers like Google and Facebook--one that IBM did not tap. Lenovo seems like an ambitious enough vcompany, and one that is dedicated in the longer term to the x86 market. IBM had been trying to get out of that market for some time, they just seemed to not be able to agree on the valuation of the unit.
@rick merritt: I wonder if Lenovo will use this heft to start competing for the big data center business at Google, Amazon, facebook, et al.
I should think so. X86 server hardware is now a commodity with commodity pricing, and you have to sell huge numbers to make any money. Big data center outfits like Google and Facebook are logical potential customers.
But they'll need to be able to customize: Facebook, for example, has specific ideas of what they need in a rack server that an off-the-shelf model might not match.
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.