This article is more than 1 year old
Meta hires network chip guru from Intel: What does this mean for future silicon?
Why be a customer when you can develop your own custom semiconductors
Analysis Here's something that should raise eyebrows in the datacenter world: Facebook parent company Meta has hired a veteran networking chip engineer from Intel to lead silicon design efforts in the internet giant's infrastructure hardware engineering group.
Jon Dama started as director of silicon in May for Meta's infrastructure hardware group, a role that has him "responsible for several design teams innovating the datacenter for scale," according to his LinkedIn profile. In a blurb, Dama indicated that a team is already in place at Meta, and he hopes to "scale the next several doublings of data processing" with them.
Though we couldn't confirm it, we think it's likely that Dama is reporting to Alexis Bjorlin, Meta's vice president of infrastructure hardware who previously worked with Dama when she was general manager of Intel's Connectivity group before serving a two-year stint at Broadcom.
Unfortunately, Dama didn't reveal more about his role beyond the basic details, though we do suspect Meta has likely tapped his expertise to address some of the internet giant's biggest datacenter needs, including boosting its recommender engines and its hype-filled "metaverse" ambitions.
Dama helped lead Intel's IPU development with Google
At Intel, Dama helped lead silicon efforts for a variety of networking hardware for the datacenter and telecom markets. This includes the x86 giant's nascent infrastructure processing units (IPU), which belong to a growing category of chips, also known as SmartNICs and data processing units, which accelerate input/output-intensive workloads, offloaded from the host CPU cores, for networking, storage, and security.
One tidbit that Meta likely found appealing about Dama is the fact that he helped guide Intel's co-development with Google on the Mount Evans IPU ASIC, which aims to provide maximum performance in so-called hyperscale datacenter environments. During his two-year involvement on this, his title was director of cloud and IPU silicon within the CTO office of Intel's Connectivity Group.
Most recently, Dama served as executive director of silicon engineering for the Connectivity Group. It was in this role that Dama was responsible for the "full lifecycle of silicon development" for products in the group, which now falls within Intel's Network and Edge Group business unit.
In this role, Dama spent a significant amount of time on the Tofino 2 and Tofino 3 switch ASICs, which are derived from Intel's 2019 acquisition of Barefoot Networks. He also led development of "new programmable pipeline logic and Ethernet IP" with a focus on improving power efficiency.
Another potential point of interest for Meta is how Dama helped triple the size of his silicon engineering team in Intel's Connectivity Group, according to his LinkedIn page. As the group's leader, Dama said he fostered a "culture of excellence," which enabled the team to develop two projects in parallel and move to a "customer-owned tooling" model.
Dama spent more than 10 years at Intel, which he joined through its 2011 acquisition of networking ASIC maker Fulcrum Microsystems.
What Meta could use Dama's expertise for
So this brings us back to Meta. Why did the internet giant hire a veteran silicon engineer whose specialty revolves around various networking chip technologies?
We don't expect Meta to dish on the implications of Dama's role, though the web titan has been a little open in the past several months about how it is reacting to changing hardware requirements. There also have been some reports detailing Meta's recent custom silicon efforts.
Most telling perhaps was a December 2021 article by industry analyst Patrick Moorhead, who interviewed Meta executive Jason Taylor about why the company hired Bjorlin to head its infrastructure hardware group.
At the time, Taylor said that while the primary focus for Meta's silicon strategy is working with the likes of Intel, Nvidia, AMD, and Broadcom, the company is exploring the creation of custom silicon for things like memory and networking bandwidth to improve the performance of recommender engines, which underpin Facebook and Meta's other properties.
Moorhead wrote in his article that Meta brought in Bjorlin to lead infrastructure hardware to "meet its rapidly growing datacenter needs," which includes using a mix of general-purpose silicon from partners as well as "in-house special purpose processors for custom workloads."
Chip makers aren't all-in on metaverse device hardware yetREAD MORE
We do have to question to what extent Meta plans to increase its reliance on custom silicon. After all, The Information reported last month that Meta is looking to "control key technologies and reduce its reliance on off-the-shelf silicon providers." And while in-house silicon development can be expensive, it can help lower costs over time — if all goes smoothly.
Another report from several months ago said that Meta was developing custom server chips, one to improve the machine-learning performance of its recommender engines and another to boost video transcoding performance for things like livestreaming.
We have to imagine that Meta is also thinking about how the right mix of silicon can help it deliver on the infrastructure needs for its lofty vision of the metaverse. On the device side, the company has reportedly struggled building custom chips for new augmented and virtual reality products, which caused it to turn to Qualcomm for smart glasses.
In a blog post from earlier this year, Meta gave us a broad view of the infrastructure needs required by metaverse applications, saying such apps "will require significant advancements in network latency, symmetrical bandwidth and overall speed of networks." Interestingly, Meta said such work will require "industry-wide collaboration."
So we have some ambiguity on how much Meta plans to turn to custom silicon versus silicon from other companies in the future. It is likely to use a mix of its own chips for accelerating certain crucial tasks while relying on mainstream parts for general compute, storage, and the like.
We do know that Dama already has experience working on silicon that meets the specific needs of hyperscalers, which could play into new homegrown silicon efforts for Meta or co-development projects like the one Intel has with Google. ®