Industry News

Samsung Electronics Co., Ltd. today announced the PM1743 SSD for enterprise servers, integrating the PCIe (Peripheral Component Interconnect Express) 5.0 interface with Samsung’s advanced sixth-generation V-NAND.

SK hynix announced that it has begun sampling out its first 24Gb DDR5 samples to industry partners. The move comes more than a year after it released its own take of the memory standard in October 2020.

Current desktop, laptop, server and gaming computing platforms leverage PCIe 4.0 (8 GT/s) infrastructure. Looking forward, chipsets solutions from Intel, AMD, Nvidia and others feature native PCIe 5.0 (16 GT/s) support. Also, PCIe 5.0 capable motherboards, SSDs and other components are popping up in the market.

PCIe performance never slows. PCI-SIG^®, the industry association committed to advancing PCIe technology, is developing the latest PCIe 6.0^® specification.

Taiwan is looking at cooperating with three Eastern European countries on semiconductors, a minister said on Thursday, a move likely to find favour in Brussels which has been courting Taiwanese semiconductor firms to manufacture in the bloc.

With this, CXL has emerged as the protocol of choice for linking CPU memory to the memory of accelerators and, interestingly enough, as the protocol that will be used to link any kind of compute engine to external memory resources.

Now, as noted above, DDR6 consumer memory is still a very long way from becoming a reality. Following a presentation from Samsung, however, it is expected that the memory modules will feature a base transfer speed of 12,000 Mbit/s with it potentially being able to go as high as 17000 Mbit/s.

Enter DDR6, the next-generation memory standard is said to be already in development and will replace DDR5 in the future. Since DDR5 only just launched, we shouldn’t expect DDR6 till at least 2025-2026+. The DDR4 memory standard stayed with us for at least 6 years so we should expect the same timeframe for DDR6’s launch.

“We were assuming that in the next 10 years, artificial intelligence would deliver, in the enterprise software space, a market cap opportunity of $30 trillion,” the star stockpicker said at a Milken Institute conference last month. “Our new number is $80 trillion.”

FADU’s preliminary specification for its Gen 5 SSD includes sequential write speeds of up to 10.4 GB/s, and sequential read speeds of up to 14.6GB/s, all while consuming less than 5.2W of power. It also delivers up to 3.5 million random read IOPS and 734,000 random write IOPS.

KIOXIA is today announcing production ready availability of the EDSFF form factor PCIe 4.0 XD6 which follows along the lines of the M.2 SSD and is an E1.S SSD that supports 15mm and 25mm sizes. The XD6 supports 1DWPD, utilizes the 1.3c NVMe protocol, is available in 1.92 -3.84TB, and is intended for high performance and high efficiency server use and storage.

TSMC has introduced a brand-new manufacturing technology roughly every two years over the past decade. Yet as the complexity of developing new fabrication processes is compounding, it is getting increasingly difficult to maintain such a cadence. The company has previously acknowledged that it will start producing chips using its N3 (3 nm) node about four months later than the industry is used to (i.e., Q2), and in a recent conference call with analysts, TSMC revealed additional details about its latest process technology roadmap, focusing on their N3, N3E, and N2 (2 nm) technologies.

As 400G optical transceivers have been put into use, data center connectivities are moving to 800G Ethernet gradually. Recently, the OFC 2020 conference took place in San Diego focused on the 800G and more enabling technologies for 800G links as well as future fiber optics, making the 800G Ethernet become another hotspot after the 400G networks. However, 800G optics is on the stage of standards perfection and testing, there is still a long way to go for its deployment on a large scale.

With the massive proliferation of data-heavy services, including high-resolution video streaming and conferencing, cloud services infrastructure growth in 2021 is expected to reach a 27% CAGR. Consequently, while 400 gigabit ethernet (GbE) is currently enjoying widespread deployment, 800 GbE is poised to rapidly follow to address these bandwidth demands.

With the growth of 5G data traffic and AI computing, data centers need faster connectivity to meet the increasing bandwidth. High speed I/O speed beyond 112 Gb/s per lane is required. If we follow the SerDes technology revolution by doubling the data rate per lane in every 2-3 years, the next generation I/O data rate will be 224 Gb/s. This article explores options, technical challenges, and potential solutions to achieve 224 Gb/s per lane.