Search Results
keyword：Mark Liu，
47 result(s)
2022/10/12
Current US sanctions on China have extended their reach to strike at HPC and sectors such as aerospace, automotive market, and military industry TrendForce indicates, the market for high-end computing chips (including CPU, GPU, etc) has borne the brunt of these restrictions at this stage, while those providing related storage such as DRAM and NAND Flash also face potential supply disruption At present, this not only includes domestic companies in mainland China but also extends to related US-based suppliers Among them, server companies that rely on high-intensity computing will face greater scrutiny
Impact analysis on server terminal shipments
In terms of server terminal shipments, since relevant component suppliers have not yet been able to confirm whether services provided by the four major cloud service providers (CSPs) in China, Baidu, ByteDance, Alibaba, and Tencent, involve military use, before CSPs sign MOUs (memoranda of understanding), component manufacturers may temporarily delay shipments to the Chinese market However, TrendForce believes, due to the fact that current CSP buyers’ component inventories remain sufficient, the short-term impact on global server market shipment performance is relatively low and long-term impact depends on the evolution of the US Department of Commerce's rules
Huawei and Sugon, two companies that have received attention at this stage due to the US ban, have previously withdrawn from the x86 server market and turned into cloud business providers and whole server delivery has been transferred to other domestic OEMs and outsourced computing power leasing, so as not to be affected by sanctions However, due to the previous CPU ban, Sugon has turned to AMD to obtain authorization for localized chips, which may be significantly curtailed by this ban In 2022, Sugon's market share in the overall server market will be approximately 23% and 85% of the Chinese market
TrendForce believes, it cannot be ruled out that relevant Chinese OEMs may have server products that may be rendered to government supercomputing centers in the future Inspur, H3C, and Lenovo will face more exacting future scrutiny and, if consequences intensify, the mainland Chinese industrial chain may feel direct effects Although commercial servers are not currently on the list of directly restricted items, if friction between the United States and China intensifies in the future, it cannot be ruled out that the US Department of Commerce will add more potentially risky Chinese server OEMs and CSPs onto the UVL list If certification cannot be realized within 60 days of being included in the UVL list, these entities will be included on the entity list The worst case scenario will be a future trend of negative growth in Chinese server demand
Since the restrictions enumerated in this ban are primarily concentrated in the HPC field, the greatest factor affecting Sugon is the company largely providing server OEM to government departments including in supercomputers, military aerospace, and government server farms At present, there are 8 national-level supercomputing centers in mainland China and the supercomputer located in the center of Wuxi is the headquarters of China's self-developed chips including the self-developed Sunway TaihuLight As the US Department of Commerce continues to strengthen its sanctions, China's supercomputing technology and domestic research capabilities will be severely damaged in the future
Impact analysis on GPU and CPU sectors
At present, companies utilizing high-end graphics cards are primarily concentrated in the HPC sector In terms of CSPs, Alibaba and Baidu are the largest companies in mainland China These two CSP companies account for up to 60% of the market share of GPU usage in China Before the previous ban at the end of August, Chinese CSP operators had to submit purchase applications before procurement but they could not apply at all after the ban However, based on the premise that buyer inventory levels on hand remain high and the supply of goods through distribution channels is sufficient, no effect on demand is forecast until 1H23 Nonetheless, it will be a challenge in the long-term Since the ban expressly prohibits supercomputing center applications such as HPC, TrendForce assesses that GPU servers used by supercomputing centers will be directly affected, which accounts for up to 30% of China's GPU market
In terms of chip computing performance control, ECCNs 3A090 and 4A090 are newly added sanctioned items and chips with a total processing performance of more than 4,800 (inclusive) calculated by TOPS will be restricted GPUs are usually used to directly assist in performing complex operations Basically, NVIDIA's A100 PCIe Gen4 and AMD's MI250 OAM Module exceed the 4,800 limit With new high computing performance products restricted in the future, development of server acceleration computing in China will take a hit
However, the computing performance of most server CPU products is generally lower than the provisions of the ban Only Chinese-made chips such as Tianjin Haiguang face direct restrictions and other CPUs such as Intel and AMD servers will not be subject to prohibition At this stage, Intel and AMD will sign MOUs with relevant mainland Chinese manufacturers to ensure that related products cannot be used in military and supercomputing fields before shipment In today's server CPUs, the computing performance of the commonly used Intel Ice Lake CPU series does not reach the limit imposed by US sanctions
Impact analysis on the memory sector
At present, Samsung and SK hynix have also suspended their supply of product to Sugon If Sugon can clarify procured memory is not used for supercomputing, domestic server products, etc, the parties will be able to reach a consensus for shipment In the long run, Korean companies are evaluating whether they need a written commitment from each customer to disavow using purchasing memory products in supercomputers Therefore, some memory shipments may be affected before documents are signed The industry generally believes that market inventory remains relatively abundant and there will be no substantial damage to the market in the short term As far as SSD is concerned, the greatest utilization remains in the category of AI/DL (Deep Learning), since most of the data trained from DL must be stored in faster and more convenient SSDs for use in inference scenarios If the suspension of shipments caused by the current ban cannot be rectified by relevant buyer agreements, the development of Chinese server manufacturers in related AI/DL fields may be hamstrung and a calamitous decline in the market penetration rate of enterprise SSDs from international manufacturers cannot be ruled out
Impact analysis on the networking sector
There are three reasons for a relatively minor impact assessment on the well-connected suppliers in the networking sector First, there are numerous networking suppliers and many of them are in China Since the demand for key components is relatively small, Chinese suppliers should be able to keep up Second, the mainstream process in this field is a mature process and future expansion is less restricted Third, from the perspective of supplier shipments, after foundry assembly, packaging, and testing, there are multiple distribution channels for the circulation of the final product and it will be difficult to determine whether terminals are military use However, from the perspective of long-term impact, there is a high probability that Chinese manufacturers will give priority to China's local supply chain in the future to ensure future supply This move will undoubtedly deepen the resistance of other suppliers' shipments to China, so it is necessary to open up multiple shipping channels to stabilize market share
For more information on reports and market data from TrendForce’s Department of Semiconductor Research, please click here, or email Ms Latte Chung from the Sales Department at lattechung@trendforcecom
For additional insights from TrendForce analysts on the latest tech industry news, trends, and forecasts, please visit our blog at https://insidertrendforcecom/
2022/10/11
According to TrendForce's latest server-related report, the original goal of CXL (Compute Express Link) was to integrate performance between various xPUs and thereby optimize hardware costs required for AI and HPC, breaking through original hardware limitations CXL support remains sourced to the CPU and, since the server CPUs that support CXL functionality such as Intel Sapphire Rapids and AMD Genoa only support the CXL 11 specification at this time, the product that this specification can realize first is CXL Memory Expander Therefore, TrendForce believes, among various CXL-related products, CXL Memory Expander will become a precursor product and this product is also the most closely related to DRAM
Assisted by CXL memory pooling functionality, AI and HPC expected to surmount hardware limitations and drive whole server DRAM consumption
However, CXL 11 is currently only practical as memory expansion in terms of DRAM and memory pooling will not be implemented until CXL 20 At this stage, average RDIMM usage of a dual-socket CPU server is approximately 10-12 units (maximum 16 units), meaning the number of RDIMM slots and channels are not being fully utilized Thus, general computing servers still have room to upgrade their original RDIMM in the short term and only applications that require AI and HPC will have CXL requirements According to TrendForce research, the consumption of CXL memory expanders has a limited impact on the overall DRAM market and CXL memory expanders are products created to optimize HPC performance in the short term
However, under existing applications, a portion of DRAM in whole servers will still idle and foster unused capacity during operation, leading to the data center industry paying for the cost of excess DRAM In the long run, the average DRAM capacity of whole servers will increase year by year as applications become more diverse and complicate However, if CXL memory pooling becomes practical in the future, internal xPU memory resources will be utilized effectively TrendForce believes that CXL memory pooling functionality will reduce the demand for RDIMM modules purchased by buyers, which will slow the growth rate of installed server DRAM capacity in individual servers in coming years
The CXL Consortium ultimately hopes to use this interface to effectively utilize the resources of every device, thereby breaking through AI and HPC hardware bottlenecks With the assistance of CXL, the development of AI and HPC will accelerate according to model complexity and contribute to the shipment volume of related models Therefore, from this perspective, CXL will drive the average capacity of DRAM at the whole server level (calculated as combined RDIMM and CXL memory expander) However, in terms of the annual growth rate of DRAM consumption on servers, growth will slow since CXL will efficiently use DRAM installed in whole devices
Buyers who will desire a substantial amount of CXL functionality are mainly focused on high-end computing machinery, so cloud service providers will be major adopters TrendForce has also observed that some OEMs ship models to their HPC computing customers requiring large-capacity DRAM expansion, which will also create potential adopters of this product
Montage, Marvell, Microchip revenue expected to ascend again due to the rise of CXL
At present, the CXL memory expander developed by manufacturers employs DDR5 but remains limited by the speed of the PCIe 50 interface at this stage and output speed is only marginally equivalent to DDR4 DDR5 will be able to realize its full speed after CPUs support PCIe 60 or higher specifications in the future From the perspective of CXL memory expander structure, a CXL controller is required in addition to DRAM CXL controller manufacturers include Montage, Marvell, Microchip, etc Therefore, the rise of CXL not only directly drives controller supplier revenue, but also does not rule out a self-development model similar to that of module houses or cloud service providers may appear in the future to produce CXL memory expanders after preparing controllers and DRAM
To sum up, the stagnation in current server performance is expected to be improved due to the development of CXL, which will effectively increase the usage of DRAM in servers while avoiding a spike in idling costs In the future, the CXL20 specification will reform the existing hardware bottleneck and, with the assistance of memory pooling, CXL will be able to exhibit greater advantages As applications become more diverse and complex, high-intensity operations such as HPC and AI will rely on xPU more than ever With shared memory pooling, model design can break free of hardware bottlenecks and continue to build more complex architectures In addition, the introduction of CXL will be popularized on the strength of future functions, especially in the large-scale introduction of cloud services into the industry This specification can better optimize communication between servers because CXL establishes high-speed communications interconnectivity and these interactions help to expand the application of computing power between parallel servers and optimizes the total cost of ownership (TCO)
For more information on reports and market data from TrendForce’s Department of Semiconductor Research, please click here, or email Ms Latte Chung from the Sales Department at lattechung@trendforcecom
For additional insights from TrendForce analysts on the latest tech industry news, trends, and forecasts, please visit our blog at https://insidertrendforcecom/
2022/07/04
According to the latest TrendForce research, despite the rapid weakening of overall consumer demand in 1H22, DRAM manufacturers previously presented a tough stance on price negotiations and gave little ground, steadily conveying inventory pressure from buyers to sellers Facing uncertain peak-season demand in 2H22, some DRAM suppliers have begun effectively expressing clear intentions to cut prices, especially in the server field, where demand is relatively stable, in order to reduce inventory pressure This situation will cause 3Q22 DRAM pricing to drop from the previous 3~8% to nearly 10% QoQ If a price war is incited due to companies competing for sales, the drop in prices may exceed 10%
PC OEMs have continuously downgraded their shipment prospects With average DRAM inventory levels at more than two months, unless a huge price incentive exists, there is no urgent demand for procurement At the same time, thanks to continuous adoption of the advanced 1Z/1alpha process, supply continued to increase in 3Q22 DDR4 output could not be effectively reduced due to high pricing limiting the penetration rate of DDR5 The price decline of PC DRAM in 3Q22 is revised to 5~10%
At present, server DRAM inventory clients have on hand is approximately 7 to 8 weeks and the buyers’ consensus is that the price of DRAM will continue to fall due to increased inventory pressure on sellers If manufacturers are willing to provide attractive quotations, buyers are willing to discuss the possibility of volume commitments As advanced manufacturing processes progress dynamically and terminal consumer products continue to weaken, server DRAM has become the only effective sales outlet Therefore, Korean manufacturers were the first to signal a willingness to discuss a quarterly pricing reduction of more than 5%, which expanded the decline of server DRAM to 5~10% in 3Q22
Smartphone production targets continued to be downgraded due to the sluggish economy In addition, smartphone brands are also pessimistic regarding future prospects, casting a gloom over their attitude towards materials stocking However, the output of mobile DRAM still increased in 3Q22 due to the adoption of advanced processes among several manufacturers, increasing pressure on suppliers This increases the willingness of sellers to offer price concessions Given the polarized disparity between supply and demand, the pricing decline of mobile DRAM is forecast to expand to 8-13% this quarter
Demand for graphics DRAM procurement has weakened due to inflation-related reduction in consumer products demand and a faltering cryptocurrency market At the same time, migrating graphics DRAM production capacity to other types of DRAM products in not as easy as migrating standard DRAM (Commodity DRAM) capacity Although demand has weakened, it is difficult for suppliers to quickly adjust output Therefore, they are also facing increasing inventory pressure Graphics DRAM pricing in 3Q22 is revised to decline 3-8% QoQ
TV shipments lead the fall and demand related to networking and industrial applications have also shown signs of weakening The price of DDR3 is currently at a relatively high point and there is plenty of room for pricing to fall in the future Weak stocking momentum originating from DDR4-related applications does not rule out the possibility of a wider decline In addition to the expansion of output due to the introduction of advanced processes, Chinese and Taiwanese manufacturers will still add new production capacity in 2H22 The decline in consumer DRAM pricing is forecast to deepen to 8~13% this quarter
For more information on reports and market data from TrendForce’s Department of Semiconductor Research, please click here, or email Ms Latte Chung from the Sales Department at lattechung@trendforcecom
For additional insights from TrendForce analysts on the latest tech industry news, trends, and forecasts, please visit our blog at https://insidertrendforcecom/
The Webinar, ”COMPUFORUM 2022 Webinar: Diverse Market Opportunities Generated by Applications of Cyber-Physical Systems” will start from June 15, 2022 The TrendForce research team including Mark Liu, Caron Ju, Bryan Ao, and Alex Chen will present speeches regarding memory, servers, and datacenters, respectively
【Info】
When：Wednesday, June, 15
Where：COMPUFORUM 2022 Webinar (Free to access)
2022/06/28
According to TrendForce research, observing recent server market dynamics, ODM’s prior production plans have begun to gradually cool Since the material mismatch cycle has improved significantly, server motherboard suppliers' stocking momentum began falling off in 2Q22 At the same time, pandemic lockdowns in Shanghai have impacted the production of some ODMs Particularly, enterprise orders led by Inventec have borne the brunt Production plans including those of Dell and HPE have been significantly delayed but overall shipment performance will not be affected in the short term Global server shipments are still forecast to grow by 65% QoQ in 3Q22, mainly due to continued support from demand generated by companies accelerating cloud migration post-pandemic
TrendForce indicates that there is currently no sign of the four major North American CSPs reducing server order volume However, since the advent of the pandemic, the industry has faced supply chain issues and the server side has continued to be affected by a limited supply of scarce materials In order to achieve production goals, buyers have increased their orders and raised their material inventories as a bulwark against shortages causing an inability to ship whole devices This prompted data centers, OEM clients, and ODMs of all sizes to increase order volume estimates, resulting in a surplus of orders in the server market Therefore, TrendForce does not rule out the possibility of subsequent order adjustments by CSP players, which will lead to a slight downward revision in server shipments in 2022
In addition to the potential quandary of impending order adjustments foreshadowed by the overall market, server demand in the Chinese market is characterized by additional policy-related factors TrendForce indicates, since the implementation of the Chinese government's policy on energy consumption and Internet business in 2021, first-tier Internet service providers in mainland China have begun to adjust their server stocking plans in 2022 Up until now, Baidu, Alibaba, and Tencent have all lowered their procurement scale this year, with Tencent downgrading its purchasing in the most obvious fashion In addition, the three major BAT giants will be more reserved in terms of future cloud construction It is worth mentioning that among the four major CSP players in China, only ByteDance has grown this year despite prevailing trends ByteDance reduces the negative impact of domestic policies in China by actively expanding its e-commerce platform that combines its overseas TikTok business and new businesses ByteDance will still maintain an annual growth rate of nearly 70% in server purchases in 2022
However, a project to shift computing resources westward on the domestic front proposed by the Chinese government is driving the construction of forward-looking servers and began priming the cloud business of telecom operators in 2Q22 TrendForce observed, in addition to China Mobile, which previously displayed strong momentum, starting to gradually increase order volume, China Telecom and China Unicom have also generally increased the scale of their server tenders, and such momentum has been evenly distributed among OEMs in mainland China Therefore, despite the slowdown in the purchasing power of China's first-tier CSP players, the deployment of provincial government servers and the construction of telecommunications companies supported by state-owned assets have formed a demand backstop for China's server market this year
Global server market will maintain positive growth in 2022, shipments expected to grow by 5% annually
TrendForce believes, the demand side of the global server market has grown significantly in the past two years due to the impact of the pandemic but the pandemic has also caused chaos in the supply chain and logistics, causing uncertainty in industry development Although server demand in 2H22 is subject to downside risks due to the consequences of material mismatches, the annual growth rate of server shipments for the entire year can still reach approximately 5% Instead, it may be affected by inventory adjustments and overall economic downturn in 2023 IT capital investment is likely to slow and growth will not be as strong as in the past two years
For more information on reports and market data from TrendForce’s Department of Semiconductor Research, please click here, or email Ms Latte Chung from the Sales Department at lattechung@trendforcecom
For additional insights from TrendForce analysts on the latest tech industry news, trends, and forecasts, please visit our blog at https://insidertrendforcecom/
The Webinar, ”COMPUFORUM 2022 Webinar: Diverse Market Opportunities Generated by Applications of Cyber-Physical Systems” will start from June 15, 2022 The TrendForce research team including Mark Liu, Caron Ju, Bryan Ao, and Alex Chen will present speeches regarding memory, servers, and datacenters, respectively
【Info】
When：Wednesday, June, 15
Where：COMPUFORUM 2022 Webinar (Free to access)
2022/06/15
TrendForce is hosting the online seminar “Compuforum 2022: Diverse Market Opportunities Generated by Applications of Cyber-Physical Systems” on the 15th this month, where analysts of the market intelligence firm are invited to delve into industrial tendencies pertaining to memory, server, and data center
New Cloud Era Sculpted by Global Hyperscale Establishment
Regarding the establishment of global hyperscale data centers, the alterations of industrial structure and the sizable improvement in server computing units have yielded a development in server architectures that is contrasting to traditional application services, and further propelled server demand for the industry Among which, the attainment of data centers has become the major drive for DRAM demand in recent years, occupying more than 30% of annual consumption over the DRAM market TrendForce estimates the overall server DRAM wafer starts to arrive at a record high ratio in 2025 at around 40%, and replace mobile DRAM as the spotlight for suppliers then
DDR5 Benefitting AI Applications alongside Next-Gen Memory Products upon Availability in 2H22
Existing memory solutions, regardless of DRAM or NAND Flash, have succumbed to the physical constrains of continuously shrinking processes, meaning that it is now even more difficult in constantly improving performance and optimizing cost DDR5 that will be introduced to the market during 2H22 will not only offer a faster data transmission, but also serves as a more convenience option for the computing architecture of AI and DL (Deep Learning) In addition, the market is seeing emergence of new memory variations, such as HBM, that are conforming to the computing requirements of HPC and AI Traditional server architectures, upon entering the AI era, are no longer able to fulfill computing demand, and are thus incorporated with accelerators in order to improve computing performance, where CXL can even optimize the interconnection between CPUs and accelerators TrendForce believes that the integration of DDR5, HBM, accelerators, and CXL should dismantle the hardware bottlenecks necessary to AI computing, and further stimulate a new wave of server growth
NAND PCIe to Elevate New Values of Data Centers from the Help of Intergenerational Computing
Storage application from the server field will become the major consumption of NAND Flash in the future, with Samsung, Solidigm, and Micron being the three prominent suppliers, and the interface will focus on EDSFF (E1S/L) in the hope of obtaining better transmission utilization that would further optimize AI and DL applications TrendForce also noticed that the QLC architecture is becoming increasingly popular owing to its effectiveness in lowering the cost unit capacity, which allows consumers to enjoy faster transmission speed with a price tag that is close to traditional HDD However, the particular architecture is not necessarily flawless as it evidently trails behind the current mainstream TLC architecture pertaining to durability, and that is the reason why DRAM and NAND will eventually establish related solutions, such as products adopted with the CXL interface, so as to achieve a balance between performance and cost In the long term, the development of storage technology will derive multiple products in order to adapt to performance or cost-oriented requirements, which at the same time erode the survival space of traditional HDD
Demand for Global IT Asset Disposition Emerges amidst Explosive Growth in Servers
The continuous progress in digital transformation among businesses, coupling with how the consumer market is now even more dependent on e-commerce and online streaming services, have further expedited the construction of global data centers The cloud computing field, due to the unabated climbing demand for AI model training, has ramped up the computability of server chips, as well as accelerated the iteration of computing platforms, where a large volume of server equipment, adopted with the previous generation computing technology, that is installed in partial institutions or CSPs with intensive data, must now be phased out, which stems out the demand for IT asset disposition (ITAD) TrendForce commented that ITRenew, Sims Lifecycle Services (SLS), and TES are the current leaders in the global ITAD market
SLS has been minimizing the area of ITAD storage facilities, expanding the cooperative network with local contractors such as waste recyclers, and lowering the maintenance cost of facilities in recent years, and turned to investing additional resources into the on-site packing services of data centers, as well as improving the flexible dispatching of resources ITRenew and TES were respectively acquired or partially acquired by information security service leaders Iron Mountain and South Korea-based SK Group during 1Q22 and 2Q22, where the ITRenew and Iron Mountain alliance will continue to provide integrated services of “software and hardware assimilation” for the ITAD market, while SK, relying on its strong capital foundation, will penetrate into the ITAD market by taking advantages of TES’ global distribution bases, where the two are anticipating that data center businesses would become robust growth dynamics for the ITAD market in the future
The “Compuforum 2022: Diverse Market Opportunities Generated by Applications of Cyber-Physical Systems” online seminar (English session only) will be available for viewing on the website of TrendForce on June 15th (Wed) TrendForce analysts Mark Liu, Bryan Ao, Caron Ju, and Alex Chen are invited to delve into various trends such as memory, server, and data center at this seminar
