Google was among the first hyperscalers build custom silicon for its services, starting first with tensor processing units (TPUs) for its AI initiatives, and then video transcoding units (VCUs) for the YouTube service. But unlike its industry peers, the company has been slower to adopt custom CPU designs, prefering to stick to off-the-shelf chips from the major CPUs. This is finally changing at Google, with the announcement that the company has developed its own in-house datacenter CPU, the Axion.
Google's Axion processor is based on the Arm Neoverse V2 (Arm v9) platform, which is Arm's current-generation design for high-performance server CPUs, and is already employed in other chips such as NVIDIA's Grace and Amazon's Graviton4. Within Google, Axion is aimed at a wide variety of workloads, including web and app servers, data analytics, microservices, and AI training. Google claims that the Axion processors boast up to 50% higher performance and up to 60% better energy efficiency compared to current-generation x86-based processors, as well as offer a 30% higher performance compared to competing Arm-based CPUs for datacenters. Though as is increasingly common for the cryptic cloud side of Google's business, least for now the company isn't specifying what processors they're comparing Axion to in these metrics.
While Google is not disclosing core counts or the full specifications of its Axion CPUs, the company is revealing that they are incorporating their own secret sauce into the silicon in the form of the company's Titanium purpose-built microcontrollers. These microcontrollers are designed to handle basic operations like networking and security, as well as offload storage I/O processing to Hyperdisk block storage service. As a result of this offloading, virtually all of the CPU core resources should be available to actual workloads. As for the chip's memory subsystem, Axion uses conventional dual-rank DDR5 memory modules.
"Google's announcement of the new Axion CPU marks a significant milestone in delivering custom silicon that is optimized for Google's infrastructure, and built on our high-performance Arm Neoverse V2 platform," said Rene Haas, CEO of Arm. "Decades of ecosystem investment, combined with Google's ongoing innovation and open-source software contributions ensure the best experience for the workloads that matter most to customers running on Arm everywhere."
Google has previously deployed Arm-based processors for its own services, including BigTable, Spanner, BigQuery, and YouTube Ads and is ready to offer instances based on its Armv9-based Axion CPUs to its customers that can use software developed for Arm architectures.
Sources: Google, Wall Street Journal
CPUsWhen Micron announced plans to build two new fabs in the U.S. in 2022, the company vaguely said both would come online by the decade's end. Then, in 2023, it began to optimize its spending, which pushed production at these fabrication facilities. This week, the company outlined more precise timeframes for when its fabs in Idaho and New York will start operations: this will happen from calendar 2026 to calendar 2029.
"These fab construction investments are necessary to support supply growth for the latter half of this decade," a statement by Micron in its Q3 FY2024 financial results report reads. "This Idaho fab will not contribute to meaningful bit supply until fiscal 2027 and the New York construction capex is not expected to contribute to bit supply growth until fiscal 2028 or later. The timing of future [wafer fab equipment] spend in these fabs will be managed to align supply growth with expected demand growth."
Micron's fiscal year 2027 starts in September 2026, so the new fab near Boise, Idaho, is set to start operations between September 2026 and September 2027. The company's fiscal 2028 starts in September 2027, so the fab will likely begin operations in calendar 2028 or later, probably depending on the demand for DRAM memory in the coming years. That said, Micron's U.S. memory fabs will begin operations between late 2026 and 2029, which aligns with the company's original plans.
Construction of the fab in Idaho is well underway. In contrast, construction of the New York facility has yet to begin as the company is working on regulatory and permitting processes in the state.
Micron's capital expenditure (CaPex) plan for FY2024 is approximately $8.0 billion, with a decrease in year-over-year spending on wafer fabrication equipment (WFE). In Q4 FY2024, the company will spend around $3 billion on fab construction, new wafer fab tools, and various expansions/upgrades.
Looking ahead to FY2025, the company plans a substantial increase in capex, targeting a mid-30s percentage of revenue to support various technological and facility advancements. In particular, it expects its quarterly CapEx to average above the $3 billion level seen in the fourth quarter of FY2024, which means that it plans to spend about $12 billion in its fiscal 2025, which begins in late September.
Half or more of the total CapEx increase in FY2025 (i.e., over $2 billion) will be allocated to constructing new fabs in Idaho and New York. Meanwhile, the FY2025 CapEx will significantly rise to fund high-bandwidth memory (HBM) assembly and testing and the construction of fabrication and back-end facilities. This increase also includes investments in technology transitions to meet growing demand.
"Fab construction in Idaho is underway, and we are working diligently to complete the regulatory and permitting processes in New York," said Sanjay Mehrotra, chief executive officer of Micron, at the company's conference call with investors and financial analysts (via SeekingAlpha). "This additional leading-edge greenfield capacity, along with continued technology transition investments in our Asia facilities, is required to meet long-term demand in the second half of this decade and beyond. These investments support our objective to maintain our current bit share over time and to grow our memory bit supply in line with long-term industry bit demand."
MemoryKioxia's booth at FMS 2024 was a busy one with multiple technology demonstrations keeping visitors occupied. A walk-through of the BiCS 8 manufacturing process was the first to grab my attention. Kioxia and Western Digital announced the sampling of BiCS 8 in March 2023. We had touched briefly upon its CMOS Bonded Array (CBA) scheme in our coverage of Kioxial's 2Tb QLC NAND device and coverage of Western Digital's 128 TB QLC enterprise SSD proof-of-concept demonstration. At Kioxia's booth, we got more insights.
Traditionally, fabrication of flash chips involved placement of the associate logic circuitry (CMOS process) around the periphery of the flash array. The process then moved on to putting the CMOS under the cell array, but the wafer development process was serialized with the CMOS logic getting fabricated first followed by the cell array on top. However, this has some challenges because the cell array requires a high-temperature processing step to ensure higher reliability that can be detrimental to the health of the CMOS logic. Thanks to recent advancements in wafer bonding techniques, the new CBA process allows the CMOS wafer and cell array wafer to be processed independently in parallel and then pieced together, as shown in the models above.
The BiCS 8 3D NAND incorporates 218 layers, compared to 112 layers in BiCS 5 and 162 layers in BiCS 6. The company decided to skip over BiCS 7 (or, rather, it was probably a short-lived generation meant as an internal test vehicle). The generation retains the four-plane charge trap structure of BiCS 6. In its TLC avatar, it is available as a 1 Tbit device. The QLC version is available in two capacities - 1 Tbit and 2 Tbit.
Kioxia also noted that while the number of layers (218) doesn't compare favorably with the latest layer counts from the competition, its lateral scaling / cell shrinkage has enabled it to be competitive in terms of bit density as well as operating speeds (3200 MT/s). For reference, the latest shipping NAND from Micron - the G9 - has 276 layers with a bit density in TLC mode of 21 Gbit/mm2, and operates at up to 3600 MT/s. However, its 232L NAND operates only up to 2400 MT/s and has a bit density of 14.6 Gbit/mm2.
It must be noted that the CBA hybrid bonding process has advantages over the current processes used by other vendors - including Micron's CMOS under array (CuA) and SK hynix's 4D PUC (periphery-under-chip) developed in the late 2010s. It is expected that other NAND vendors will also move eventually to some variant of the hybrid bonding scheme used by Kioxia.
StorageG.Skill on Tuesday introduced its ultra-low-latency DDR5-6400 memory modules that feature a CAS latency of 30 clocks, which appears to be the industry's most aggressive timings yet for DDR5-6400 sticks. The modules will be available for both AMD and Intel CPU-based systems.
With every new generation of DDR memory comes an increase in data transfer rates and an extension of relative latencies. While for the vast majority of applications, the increased bandwidth offsets the performance impact of higher timings, there are applications that favor low latencies. However, shrinking latencies is sometimes harder than increasing data transfer rates, which is why low-latency modules are rare.
Nonetheless, G.Skill has apparently managed to cherry-pick enough DDR5 memory chips and build appropriate printed circuit boards to produce DDR5-6400 modules with CL30 timings, which are substantially lower than the CL46 timings recommended by JEDEC for this speed bin. This means that while JEDEC-standard modules have an absolute latency of 14.375 ns, G.Skill's modules can boast a latency of just 9.375 ns – an approximately 35% decrease.
G.Skill's DDR5-6400 CL30 39-39-102 modules have a capacity of 16 GB and will be available in 32 GB dual-channel kits, though the company does not disclose voltages, which are likely considerably higher than those standardized by JEDEC.
The company plans to make its DDR5-6400 modules available both for AMD systems with EXPO profiles (Trident Z5 Neo RGB and Trident Z5 Royal Neo) and for Intel-powered PCs with XMP 3.0 profiles (Trident Z5 RGB and Trident Z5 Royal). For AMD AM5 systems that have a practical limitation of 6000 MT/s – 6400 MT/s for DDR5 memory (as this is roughly as fast as AMD's Infinity Fabric can operate at with a 1:1 ratio), the new modules will be particularly beneficial for AMD's Ryzen 7000 and Ryzen 9000-series processors.
G.Skill notes that since its modules are non-standard, they will not work with all systems but will operate on high-end motherboards with properly cooled CPUs.
The new ultra-low-latency memory kits will be available worldwide from G.Skill's partners starting in late August 2024. The company did not disclose the pricing of these modules, but since we are talking about premium products that boast unique specifications, they are likely to be priced accordingly.
Memory
0 Comments