By Junko Yoshida

What’s at stake:
Picture chips in data centers silently failing, leaving no trace in system logs. Such undetectable errors could steadily spread contagion across several services. Consider such a scenario in a two-engine airplane. Suppose one of the engines silently dies, unnoticed. After landing, the pilot takes off again for a new mission, assuming he has two functioning engines. One could say that this is impossible because the pilot can see — and hear — the busted engine. Unlike the plane, a busted engine in a datacenter hyperscaler can’t be seen or heard, and won’t kill anyone. But the silent crash of a critical component could trigger system-wide failures.

Designers, manufacturers and users of chips have long dreaded “soft errors,” if chips subjected to particle strikes from cosmic rays suffer unexpected bit flips.

Meanwhile, hyperscalers are lately alarmed about “hard errors” in chips with a physical defect that slipped through the manufacturing testing process or degraded gradually while deployed for a long time.

Both types of error are devastating to computing systems, especially when their “silence” affects critical missions. When chips give no indication that something has gone wrong or miscalculated, the phenomenon is called “Silent Data Corruption (SDC).”

Read More »What If Your Chip, Plane, Data Center Silently Failed?

By Bolaji Ojo

Intel Corp.’s massive challenges and the efforts of its recent leaders to thread their way back to growth have become the subject of MBA studies.

Additional fodder for business school forensic examination of Intel cropped up last week in CEO Patrick Gelsinger’s subtle but intriguing hints about the chipmaker’s future.

After repeatedly insisting that the microprocessor supplier’s revitalization plans do not include a breakup, Gelsinger last week said: “We’re building two world-class companies,” referring to the company’s products group and Intel Foundry Services, the unit created to provide outsourced semiconductor manufacturing services to other chipmakers.

To further deepen the mystery, Gelsinger said his goal is to turn the company’s products group into a “world-class fabless company.” Will these two entities co-exist under the same stable or will they separate and fashion their separate futures as independent entities?

Read More »Intel Says it’s Building ‘Two World-Class Companies.’ Meaning, please?

By Junko Yoshida

What’s at stake:
To Cuda or not to Cuda revives the decades-old debate over licensed vs. open-source software. But at stake is the safety, for those who choose to develop their own AV software stack outside the context of a safety-certified Nvidia’s SoC and Drive OS. The onus of qualification is now placed on the carmakers’ system integrators. It’s “a huge undertaking,” to say the least, according to a safety expert.

In designing next-generation highly automated vehicles, carmakers’ top priority has to be the right advanced automotive SoC. OEMs need a highly integrated chip with enough processing capability to power neural networks, support sensor fusion and manage central engine functions in new ADAS models.

Read More »With Nvidia, It’s Always Take It or Leave It

By Ron Wilson

What’s at stake:
2 nm process nodes will require novel transistors and will push the limits of EUV lithography. But if the interconnect can’t scale as well as the transistors, the new processes will deliver neither the speed, nor the density, nor the power savings designers are seeking.

As the leading edge of the semiconductor manufacturing industry — that is, Intel, Samsung, and TSMC — grinds inexorably toward the 2 nm process node, there has been much discussion of new kinds of transistors and of the new demands on EUV lithography. But another element of process technology is equally critical to the success of 2nm, and equally hard to scale: the interconnect wires that connect the transistors into circuits, and the circuits into functional blocks.

Read More »As 2nm Approaches, the Focus Shifts Toward Interconnect

By Junko Yoshida

CrowdStrike released a preliminary incident report on the catastrophic software update that caused a global IT outage last Friday.

The company’s proposed remedies, in a section entitled “How Do We Prevent This From Happening Again?” parallel recommendations in our previous story. 

CrowdStrike now says it will “implement a staggered deployment strategy for Rapid Response Content.” But some details in the preliminary report are surprising, particularly CrowdStrike’s explanation of how it implemented massively deploy its so-called “Rapid Response Content” without testing.

It turns out, as professor Phil Koopman of Carnegie Mellon University, concluded in our recent interview, that Crowdstrike tests software changes subject to phased release by IT groups. “But it pushes data changes straight to production with NO TESTING. The only precaution is a check by CrowdStrike’s own Content Validator.” The Content Validator was defective, added Koopman, “allowing bad content to get through.”

Read More »Update: CrowdStrike Pushed ‘Data Changes’ Without Testing