As sales of GPU-based AI accelerators remain as strong as ever, the immense demand for these cards has led to some server builders going off the beaten path in order to get the hardware they want at a lower price. While both NVIDIA and AMD offer official card configurations for servers, the correspondingly high price of these cards makes them a significant financial outlay that some customers either can't afford, or don't want to pay.
Instead, these groups have been turning to buying up consumer graphics cards, which although they come with additional limitations, are also a fraction of the cost of a "proper" server card. And this week, ASRock has removed another one of those limitations for would-be AMD Radeon users, with the introduction of a set of compact, passively-cooled Radeon RX 7900 XTX and RX 7900 XT video cards that are designed to go in servers.
Without any doubts, ASRock's AMD Radeon RX 7900 XTX Passive 24GB and AMD Radeon RX 7900 XT Passive 20GB AIBs are indeed graphics cards with four display outputs and based on the Navi 31 graphics processor (with 6144 and 5376 stream processors, respectively), so they can output graphics and work both with games and professional applications. And with TGPs of 355W and 315W respectively, these cards aren't underclocked in any way compared to traditional desktop cards. However, unlike a typical desktop card, the cooler on these cards is a dual-slot heatsink without any kind of fan attached, which is meant to be used with high-airflow forced-air cooling.
All-told, ASRock's passive cooler is pretty capable, as well; it's not just a simple aluminum heatsink. Beneath the fins, ASRock has gone with a vapor chamber and multiple heat pipes to distribute heat to the rest of the sink. Even with forced-air cooling in racked servers, the heatsink itself still needs to be efficient to keep a 300W+ card cool with only a dual-slot cooler – and especially so when upwards of four of these cards are installed side-by-side with each other. To make the boards even more server friendly, these cards are equipped with a 12V-2×6 power connector, a first for the Radeon RX 7900 series, simplifying installation by reducing cable clutter.
Driving the demand for these cards in particular is their memory configuration. With 24GB for the 7900 XTX and 20GB for the 7900 XT is half as much (or less) memory than can be found on AMD and NVIDIA's high-end professional and server cards, AMD is the only vendor offering consumer cards with this much memory for less than $1000. So for a memory-intensive AI inference cluster built on a low budget, the cheapest 24GB card available starts looking like a tantalizing option.
Otherwise, ASRock's Radeon RX 7900 Passive cards distinguish themselves from AMD's formal professional and server cards by what they're not capable of doing: namely, remote professional graphics or other applications that need things like GPU partitioning. These parts look to be aimed at one application only, artificial intelligence, and are meant to process huge amounts of data. For this purpose, their passive coolers will do the job and the lack of ProViz or VDI-oriented drives ensure that AMD will leave these lucrative markets for itself.
GPUsWhile Realtek is best known in the enthusiast space for for its peripheral controllers such as audio codecs and network controllers, the company also has a small-but-respectable SSD controller business that tends to fly under the radar due to its focus on entry-level and mainstream drives. But Realtek's stature in the SSD space is on the rise, as the company is not only planning new PCIe Gen5 SSD controllers, but also their first high-end, DRAM-equipped SSD controller.
For this year's Computex trade show, Realtek laid out a new SSD controller roadmap that calls for the company to release a trio of new SSD controllers over the next couple of years. First up is a new four-channel entry-level PCIe 4.0 controller, the RTS5776DL, which will be joined a bit later by a PCIe 5.0 variant, the RTS5781DL. But most interesting on Realtek's new roadmap is the final chip being planned: the eight-channel, DRAM-equipped RTS5782, which would be the company's first high-end SSD controller, capable of hitting sequential read rates as high as 14GB/second.
| Realtek NVMe SSD Controller Comparison | |||||||||
| RTS5782 | RTS5781DL | RTS5776DL | RTS5772DL | RTS5766DL | |||||
| Market Segment | High-End | Mainstream | Entry-Level | ||||||
| Error Correction | 4K LDPC | 2K LDPC | |||||||
| DRAM | DDR4, LPDDR4(X) | No | No | No | No | ||||
| Host Interface | PCIe 5.0 x4 | PCIe 5.0 x4 | PCIe 4.0 x4 | PCIe 4.0 x4 | PCIe 3.0 x4 | ||||
| NVMe Version | NVMe 2.0 | NVMe 2.0 | NVMe 2.0 | NVMe 1.4 | NVMe 1.4 | ||||
| NAND Channels, Interface Speed | 8 ch, 3600 MT/s |
4 ch, 3600 MT/s |
4 ch, 3600 MT/s |
8 ch, 1600 MT/s |
4 ch, 1200 MT/s |
||||
| Sequential Read | 14 GB/s | 10 GB/s | 7.4 GB/s | 6 GB/s | 3.2 GB/s | ||||
| Sequential Write | 12 GB/s | 10 GB/s | 7.4 GB/s | 6 GB/s | 2.2 GB/s | ||||
| 4KB Random Read IOPS | 2500k | 1400k | 1200k | - | - | ||||
| 4KB Random Write IOPS | 2500k | 1400k | 1200k | - | - | ||||
Diving a bit deeper into Realtek's roadmap, the RTS5776DL is traditional DRAM-less PCIe Gen4 x4 controller with four NAND chann... SSDs
Computex keynote season is kicking into high gear this morning with the show's leading keynote, which is being delivered by AMD. Company CEO Dr. Lisa Su will be presenting a keynote entitled “The future of high-performance computing in the AI era,” and with a run time of 90 minutes, we're expecting AMD to have a whole host of product announcements covering their full spectrum of product categories.
The big expectation here is fresh news around AMD’s Zen 5 CPU core architecture, and the chips built around it. AMD’s most recent Zen 5 roadmap has it slated to deliver all three flavors of Zen 5 by the end of this year, and we’re coming up on the two-year anniversary of the Zen 4 architecture launch.
Along with client chips, AMD has been pushing their server CPUs hard, and they’ve previously told investors that the next-gen EPYC Turin CPU is “looking great”. So we’ll likely hear about both client and server Zen 5 product plans during this keynote.
On the GPU/accelerator side of matters, AMD is mid-cycle (at best) with their Instinct MI300 series accelerators. With the company’s sales repeatedly beating their own expectations, AMD doesn’t seem to need much help moving this premium silicon right now. But with AI being the operative buzzword of this year’s Computex (and indeed, the computing industry as a whole), it would be weird for AMD to not have something to say about their rapidly growing AI accelerator product line.
Come join us at 6:30pm PT / 9:30pm ET / 01:30 UTC to get all the details.
Live BlogWhen a major industry slowdown occurs, big companies tend to slowdown their mid-term and long-term capacity related investments. This is exactly what happened to SK hynix's Yongin Semiconductor Cluster, a major project announced in April 2021 and valued at $106 billion. While development of the site has been largely completed, only 35% of the initial shell building has been constructed, according to the Korean Ministry of Trade, Industry, and Energy.
"Approximately 35% of Fab 1 has been completed so far and site renovation is in smooth progress," a statement by the Korean Ministry of Trade, Industry, and Energy reads. "By 2046, over KRW 120 trillion ($90 billion today, $106 billion in 2021) in investment will be poured to complete Fabs 1 through 4, and construction of Fab 1's production line will commence in March next year. Once completed, the infrastructure will rank as the world's largest three-story fab."
The new semiconductor fabrication cluster by SK hynix announced almost exactly three years ago is primarily meant to be used to make DRAM for PCs, mobile devices, and servers using advanced extreme ultraviolet lithography (EUV) process technologies. The cluster, located near Yongin, South Korea, is intended to consist of four large fabs situated on a 4.15 million m2 site. With a planned capacity of approximately 800,000 wafer starts per month (WSPMs), it is set to be one of the world's largest semiconductor production hubs.
With that said, SK hynix's construction progress has been slower than the company first projected. The first fab in the complex was originally meant to come online in 2025, with construction starting in the fourth quarter of 2021. However, SK hynix began to cut its capital expenditures in the second half of 2022, and the Yongin Semiconductor Cluster project fell a victim of that cut. To be sure, the site continues to be developed, just at a slower pace; which is why some 35% of the first fab shell has been built at this point.
If completed as planned in 2021, the first phase of SK hynix Yongin operations would have been a major memory production facility costing $25 billion, equipped with EUV tools, and capable of 200,000-WSPM, according to reports from 2021.
Sources: Korean Ministry of Trade, Industry, and Energy; ComputerBase
MemoryWhile Realtek is best known in the enthusiast space for for its peripheral controllers such as audio codecs and network controllers, the company also has a small-but-respectable SSD controller business that tends to fly under the radar due to its focus on entry-level and mainstream drives. But Realtek's stature in the SSD space is on the rise, as the company is not only planning new PCIe Gen5 SSD controllers, but also their first high-end, DRAM-equipped SSD controller.
For this year's Computex trade show, Realtek laid out a new SSD controller roadmap that calls for the company to release a trio of new SSD controllers over the next couple of years. First up is a new four-channel entry-level PCIe 4.0 controller, the RTS5776DL, which will be joined a bit later by a PCIe 5.0 variant, the RTS5781DL. But most interesting on Realtek's new roadmap is the final chip being planned: the eight-channel, DRAM-equipped RTS5782, which would be the company's first high-end SSD controller, capable of hitting sequential read rates as high as 14GB/second.
| Realtek NVMe SSD Controller Comparison | |||||||||
| RTS5782 | RTS5781DL | RTS5776DL | RTS5772DL | RTS5766DL | |||||
| Market Segment | High-End | Mainstream | Entry-Level | ||||||
| Error Correction | 4K LDPC | 2K LDPC | |||||||
| DRAM | DDR4, LPDDR4(X) | No | No | No | No | ||||
| Host Interface | PCIe 5.0 x4 | PCIe 5.0 x4 | PCIe 4.0 x4 | PCIe 4.0 x4 | PCIe 3.0 x4 | ||||
| NVMe Version | NVMe 2.0 | NVMe 2.0 | NVMe 2.0 | NVMe 1.4 | NVMe 1.4 | ||||
| NAND Channels, Interface Speed | 8 ch, 3600 MT/s |
4 ch, 3600 MT/s |
4 ch, 3600 MT/s |
8 ch, 1600 MT/s |
4 ch, 1200 MT/s |
||||
| Sequential Read | 14 GB/s | 10 GB/s | 7.4 GB/s | 6 GB/s | 3.2 GB/s | ||||
| Sequential Write | 12 GB/s | 10 GB/s | 7.4 GB/s | 6 GB/s | 2.2 GB/s | ||||
| 4KB Random Read IOPS | 2500k | 1400k | 1200k | - | - | ||||
| 4KB Random Write IOPS | 2500k | 1400k | 1200k | - | - | ||||
Diving a bit deeper into Realtek's roadmap, the RTS5776DL is traditional DRAM-less PCIe Gen4 x4 controller with four NAND chann... SSDs
Computex keynote season is kicking into high gear this morning with the show's leading keynote, which is being delivered by AMD. Company CEO Dr. Lisa Su will be presenting a keynote entitled “The future of high-performance computing in the AI era,” and with a run time of 90 minutes, we're expecting AMD to have a whole host of product announcements covering their full spectrum of product categories.
The big expectation here is fresh news around AMD’s Zen 5 CPU core architecture, and the chips built around it. AMD’s most recent Zen 5 roadmap has it slated to deliver all three flavors of Zen 5 by the end of this year, and we’re coming up on the two-year anniversary of the Zen 4 architecture launch.
Along with client chips, AMD has been pushing their server CPUs hard, and they’ve previously told investors that the next-gen EPYC Turin CPU is “looking great”. So we’ll likely hear about both client and server Zen 5 product plans during this keynote.
On the GPU/accelerator side of matters, AMD is mid-cycle (at best) with their Instinct MI300 series accelerators. With the company’s sales repeatedly beating their own expectations, AMD doesn’t seem to need much help moving this premium silicon right now. But with AI being the operative buzzword of this year’s Computex (and indeed, the computing industry as a whole), it would be weird for AMD to not have something to say about their rapidly growing AI accelerator product line.
Come join us at 6:30pm PT / 9:30pm ET / 01:30 UTC to get all the details.
Live BlogWhen a major industry slowdown occurs, big companies tend to slowdown their mid-term and long-term capacity related investments. This is exactly what happened to SK hynix's Yongin Semiconductor Cluster, a major project announced in April 2021 and valued at $106 billion. While development of the site has been largely completed, only 35% of the initial shell building has been constructed, according to the Korean Ministry of Trade, Industry, and Energy.
"Approximately 35% of Fab 1 has been completed so far and site renovation is in smooth progress," a statement by the Korean Ministry of Trade, Industry, and Energy reads. "By 2046, over KRW 120 trillion ($90 billion today, $106 billion in 2021) in investment will be poured to complete Fabs 1 through 4, and construction of Fab 1's production line will commence in March next year. Once completed, the infrastructure will rank as the world's largest three-story fab."
The new semiconductor fabrication cluster by SK hynix announced almost exactly three years ago is primarily meant to be used to make DRAM for PCs, mobile devices, and servers using advanced extreme ultraviolet lithography (EUV) process technologies. The cluster, located near Yongin, South Korea, is intended to consist of four large fabs situated on a 4.15 million m2 site. With a planned capacity of approximately 800,000 wafer starts per month (WSPMs), it is set to be one of the world's largest semiconductor production hubs.
With that said, SK hynix's construction progress has been slower than the company first projected. The first fab in the complex was originally meant to come online in 2025, with construction starting in the fourth quarter of 2021. However, SK hynix began to cut its capital expenditures in the second half of 2022, and the Yongin Semiconductor Cluster project fell a victim of that cut. To be sure, the site continues to be developed, just at a slower pace; which is why some 35% of the first fab shell has been built at this point.
If completed as planned in 2021, the first phase of SK hynix Yongin operations would have been a major memory production facility costing $25 billion, equipped with EUV tools, and capable of 200,000-WSPM, according to reports from 2021.
Sources: Korean Ministry of Trade, Industry, and Energy; ComputerBase
Memory
In what started last year as a handful of reports about instability with Intel's Raptor Lake desktop chips has, over the last several months, grown into a much larger saga. Facing their biggest client chip instability impediment in decades, Intel has been under increasing pressure to figure out the root cause of the issue and fix it, as claims of damaged chips have stacked up and rumors have swirled amidst the silence from Intel. But, at long last, it looks like Intel's latest saga is about to reach its end, as today the company has announced that they've found the cause of the issue, and will be rolling out a microcode fix next month to resolve it.
Officially, Intel has been working to identify the cause of desktop Raptor Lake’s instability issues since at least February of this year, if not sooner. In the interim they have discovered a couple of correlating factors – telling motherboard vendors to stop using ridiculous power settings for their out-of-the-box configurations, and finding a voltage-related bug in Enhanced Thermal Velocity Boost (eTVB) – but neither factor was the smoking gun that set all of this into motion. All of which had left Intel to continue searching for the root cause in private, and lots of awkward silence to fill the gaps in the public.
But it looks like Intel’s search has finally come to an end – even if Intel isn’t putting the smoking gun on public display quite yet. According to a fresh update posted to the company’s community website, Intel has determined the root cause at last, and has a fix in the works.
Per the company’s announcement, Intel has tracked down the cause of the instability issue to “elevated operating voltages”, that at its heart, stems from a flawed algorithm in Intel’s microcode that requested the wrong voltage. Consequently, Intel will be able to resolve the issue through a new microcode update, which pending validation, is expected to be released in the middle of August.
And while there’s nothing good for Intel about Raptor Lake’s instability issues or the need to fix them, that the problem can be ascribed to (or at least fixed by) microcode is about the best possible outcome the company could hope for. Across the full spectrum of potential causes, microcode is the easiest to fix at scale – microcode updates are already distributed through OS updates, and all chips of a given stepping (millions in all) run the same microcode. Even a motherboard BIOS-related issue would be much harder to fix given the vast number of different boards out there, never mind a true hardware flaw that would require Intel to replace even more chips than they already have.
Still, we’d also be remiss if we didn’t note that microcode is regularly used to paper over issues further down in the processor, as we’ve most famously seen with the Meltdown/Spectre fixes several years ago. So while Intel is publicly attributing the issue to microcode bugs, there are several more layers to the onion that is modern CPUs that could be playing a part. In that respect, a microcode fix grants the least amoun... CPUs
At FMS 2024, the technological requirements from the storage and memory subsystem took center stage. Both SSD and controller vendors had various demonstrations touting their suitability for different stages of the AI data pipeline - ingestion, preparation, training, checkpointing, and inference. Vendors like Solidigm have different types of SSDs optimized for different stages of the pipeline. At the same time, controller vendors have taken advantage of one of the features introduced recently in the NVM Express standard - Flexible Data Placement (FDP).
FDP involves the host providing information / hints about the areas where the controller could place the incoming write data in order to reduce the write amplification. These hints are generated based on specific block sizes advertised by the device. The feature is completely backwards-compatible, with non-FDP hosts working just as before with FDP-enabled SSDs, and vice-versa.
Silicon Motion's MonTitan Gen 5 Enterprise SSD Platform was announced back in 2022. Since then, Silicon Motion has been touting the flexibility of the platform, allowing its customers to incorporate their own features as part of the customization process. This approach is common in the enterprise space, as we have seen with Marvell's Bravera SC5 SSD controller in the DapuStor SSDs and Microchip's Flashtec controllers in the Longsys FORESEE enterprise SSDs.
At FMS 2024, the company was demonstrating the advantages of flexible data placement by allowing a single QLC SSD based on their MonTitan platform to take part in different stages of the AI data pipeline while maintaining the required quality of service (minimum bandwidth) for each process. The company even has a trademarked name (PerformaShape) for the firmware feature in the controller that allows the isolation of different concurrent SSD accesses (from different stages in the AI data pipeline) to guarantee this QoS. Silicon Motion claims that this scheme will enable its customers to get the maximum write performance possible from QLC SSDs without negatively impacting the performance of other types of accesses.
Silicon Motion and Phison have market leadership in the client SSD controller market with similar approaches. However, their enterprise SSD controller marketing couldn't be more different. While Phison has gone in for a turnkey solution with their Gen 5 SSD platform (to the extent of not adopting the white label route for this generation, and instead opting to get the SSDs qualified with different cloud service providers themselves), Silicon Motion is opting for a different approach. The flexibility and customization possibilities can make platforms like the MonTitan appeal to flash array vendors.
StorageKioxia'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.
StorageWhile Realtek is best known in the enthusiast space for for its peripheral controllers such as audio codecs and network controllers, the company also has a small-but-respectable SSD controller business that tends to fly under the radar due to its focus on entry-level and mainstream drives. But Realtek's stature in the SSD space is on the rise, as the company is not only planning new PCIe Gen5 SSD controllers, but also their first high-end, DRAM-equipped SSD controller.
For this year's Computex trade show, Realtek laid out a new SSD controller roadmap that calls for the company to release a trio of new SSD controllers over the next couple of years. First up is a new four-channel entry-level PCIe 4.0 controller, the RTS5776DL, which will be joined a bit later by a PCIe 5.0 variant, the RTS5781DL. But most interesting on Realtek's new roadmap is the final chip being planned: the eight-channel, DRAM-equipped RTS5782, which would be the company's first high-end SSD controller, capable of hitting sequential read rates as high as 14GB/second.
| Realtek NVMe SSD Controller Comparison | |||||||||
| RTS5782 | RTS5781DL | RTS5776DL | RTS5772DL | RTS5766DL | |||||
| Market Segment | High-End | Mainstream | Entry-Level | ||||||
| Error Correction | 4K LDPC | 2K LDPC | |||||||
| DRAM | DDR4, LPDDR4(X) | No | No | No | No | ||||
| Host Interface | PCIe 5.0 x4 | PCIe 5.0 x4 | PCIe 4.0 x4 | PCIe 4.0 x4 | PCIe 3.0 x4 | ||||
| NVMe Version | NVMe 2.0 | NVMe 2.0 | NVMe 2.0 | NVMe 1.4 | NVMe 1.4 | ||||
| NAND Channels, Interface Speed | 8 ch, 3600 MT/s |
4 ch, 3600 MT/s |
4 ch, 3600 MT/s |
8 ch, 1600 MT/s |
4 ch, 1200 MT/s |
||||
| Sequential Read | 14 GB/s | 10 GB/s | 7.4 GB/s | 6 GB/s | 3.2 GB/s | ||||
| Sequential Write | 12 GB/s | 10 GB/s | 7.4 GB/s | 6 GB/s | 2.2 GB/s | ||||
| 4KB Random Read IOPS | 2500k | 1400k | 1200k | - | - | ||||
| 4KB Random Write IOPS | 2500k | 1400k | 1200k | - | - | ||||
Diving a bit deeper into Realtek's roadmap, the RTS5776DL is traditional DRAM-less PCIe Gen4 x4 controller with four NAND chann... SSDs
Computex keynote season is kicking into high gear this morning with the show's leading keynote, which is being delivered by AMD. Company CEO Dr. Lisa Su will be presenting a keynote entitled “The future of high-performance computing in the AI era,” and with a run time of 90 minutes, we're expecting AMD to have a whole host of product announcements covering their full spectrum of product categories.
The big expectation here is fresh news around AMD’s Zen 5 CPU core architecture, and the chips built around it. AMD’s most recent Zen 5 roadmap has it slated to deliver all three flavors of Zen 5 by the end of this year, and we’re coming up on the two-year anniversary of the Zen 4 architecture launch.
Along with client chips, AMD has been pushing their server CPUs hard, and they’ve previously told investors that the next-gen EPYC Turin CPU is “looking great”. So we’ll likely hear about both client and server Zen 5 product plans during this keynote.
On the GPU/accelerator side of matters, AMD is mid-cycle (at best) with their Instinct MI300 series accelerators. With the company’s sales repeatedly beating their own expectations, AMD doesn’t seem to need much help moving this premium silicon right now. But with AI being the operative buzzword of this year’s Computex (and indeed, the computing industry as a whole), it would be weird for AMD to not have something to say about their rapidly growing AI accelerator product line.
Come join us at 6:30pm PT / 9:30pm ET / 01:30 UTC to get all the details.
Live Blog
0 Comments