SanDisk Optimus GX Pro 8TB 8100 Gen5 SSD review: PCIe 5.0 NVMe drive delivers 14.9GB/s reads, 13.2GB/s writes, and 4,800TBW endurance, but pricing has surged to $2,149 USD / $5,799 CAD amid rising NAND costs.
The SanDisk Optimus GX Pro 8100 (rebranded from WD_Black SN8100) is a PCIe 5.0 x4 NVMe 2.0 SSD available in 1TB, 2TB, 4TB, and 8TB capacities. The 8TB model uses a double-sided PCB with four SanDisk 218-layer BiCS 8 TLC CBA 3D NAND chips and a SanDisk-branded Silicon Motion SM2508 8-channel controller, plus DDR4 DRAM. Performance specs for the 8TB version: sequential reads up to 14.9GB/s, writes up to 13.2GB/s, random read/write IOPS of 2,200K/2,400K. The drive supports TCG Opal 2.02 encryption and is rated for 4,800TBW with a five-year warranty. Lower capacities are single-sided with average active power of 6.5W read/7W write, making them suitable for laptops.
Pricing has escalated dramatically: the 2TB model now costs $549.99 on Amazon (down from $279.99), while the 8TB version is listed at $2,149 in the US and an eye-watering $5,799.99 on Amazon Canada. The reviewer notes the drive’s endurance and speed parallel a cross-Canada motorcycle ride raising awareness for PTSD and mental wellness among military and first responder veterans—a cause supported by the Rolling Barrage event.
Samsung Electronics and SK Hynix are reportedly allocating additional or flexible DRAM production capacity to server DDR5 and other conventional memory products while maintaining high-bandwidth memory (HBM) volumes already committed to customers.
MSI Slyly Shows off an Upcoming DLC AMD EPYC Venice Platform With CD182-S6091-X2 Servers and Racks
19 h ago
Intel announced a mid-generation upgrade for its Xeon 6 processor family, enabling support for DDR5-8000 RDIMMs on select Xeon 6700P SKUs starting August 2025 and adding Gen 2 MRDIMM compatibility for Xeon 6900P in Q1 2027, targeting AI workload memory bandwidth bottlenecks. In a blog post, Intel’s Srini Krishna explained that CPU memory bandwidth is a bottleneck for modern AI, so the company is leveraging the overbuilt memory controller in Granite Rapids chips to run faster RDIMMs. DDR5-8000 offers a 25% bandwidth increase over the current DDR5-6400 limit, with Intel measuring a 20% improvement in memory bandwidth and a 3–6% performance gain across workloads. Latency also improves by about 6%. However, the faster speeds apply only to 1DPC (one DIMM per channel) operation; 2DPC remains at DDR5-5200. Only select Xeon 6700P chips already qualified for DDR5-8000 MRDIMM operation will support the faster RDIMMs. For the Xeon 6900P platform, Intel is adding Gen 2 MRDIMM support but without higher memory speeds—the controllers lack headroom beyond the current 8800 MT/s Gen 1 limit. The move ensures ecosystem continuity and forward compatibility as the industry transitions to newer MRDIMM generations, allowing customers to use the same memory supplies across Xeon 6 and upcoming Xeon 6+ processors. No RDIMM speed upgrade was announced for Xeon 6900P. The upgrades help close the gap with Intel’s recently launched Xeon 6+ “Clearwater Forest” chips, which shipped with similar memory support.
Read also : IT Press Tour 68: ExaGrid Confirming its unique position on the market as a pure secondary storage player dedicated to backup By Philippe Nicolas | June 15, 2026 | News
Scaleway, fournisseur européen de cloud et d’IA, a été sélectionné par Airbus comme partenaire cloud souverain pour moderniser ses applications critiques dans un environnement sécurisé et juridiquement protégé, répondant ainsi à la recherche de solutions cloud haute performance, souveraines et adaptées à l’intelligence artificielle pour les industries stratégiques.
Dans le cadre de ce contrat, Scaleway déploiera une infrastructure cloud souveraine, basée sur des technologies ouvertes et une interopérabilité native, pour héberger des charges de travail exigeant une gouvernance renforcée, une résilience maximale et une protection contre les législations extraterritoriales non européennes. Airbus a évalué plusieurs fournisseurs selon trois critères : capacités technologiques (services cloud avancés, IA, interopérabilité, montée en charge), excellence opérationnelle (sécurité, résilience, continuité de service, intégration multicloud) et garanties juridiques (juridiction européenne, protection des données).
Cette plateforme prendra en charge des applications critiques couvrant la conception d’aéronefs, l’ingénierie, la production industrielle et les opérations de l’entreprise, renforçant la protection de la propriété intellectuelle et des données industrielles d’Airbus tout en assurant la continuité de ses opérations mondiales.
« L’IA redéfinit la manière dont les industries avancées conçoivent et produisent. Nous sommes fiers qu’Airbus ait choisi Scaleway pour démontrer que l’Europe propose un cloud souverain aux standards internationaux », déclare Damien Lucas, CEO de Scaleway. Catherine Jestin, EVP Digital d’Airbus, ajoute : « Cette collaboration marque une étape clé pour notre souveraineté numérique, en intégrant un cloud de confiance protégeant nos données critiques des législations extraterritoriales. »
Cette initiative s’inscrit dans la stratégie de souveraineté numérique d’Airbus, complétant son approche multicloud pour déployer chaque charge de travail dans l’environnement le plus adapté. Scaleway et Airbus contribuent ainsi à bâtir les fondations numériques de la prochaine génération de l’aéronautique européenne, alliant innovation, performance et maîtrise des technologies critiques.
Longsys Gen5 2TB mSSD Review 13GB/s Speed up to 8TB Capacity in a 25 Cent Piece Sized Storage Device
1 d ago
Exascend, a boutique NAND flash storage manufacturer, has launched its PD5 Gen5 7.68TB Enterprise SSD, a PCIe 5.0 x4 drive designed for AI/ML, HPC, and data-intensive workloads, featuring up to 14.1GB/s sequential reads and 3.12 million IOPS. The company, which The SSD Review met at Computex in Taiwan and later visited at its engineering team in Shanghai, differentiates itself through a vertically integrated approach—developing solutions from the wafer level with in-house hardware, firmware, and software expertise to deliver highly customized, customer-focused storage rather than one-size-fits-all products.
The PD5 arrives in a U.2 form factor with a ribbed black aluminum casing and is available in capacities from 7.68TB to 30.72TB. It is rated for sequential write speeds up to 9.35GB/s and supports a broad range of enterprise operating systems including Windows Server, Red Hat Enterprise Linux, Ubuntu, VMware ESXi, and KVM. Enterprise-class features include advanced LDPC error correction, Self-Encrypting Drive (SED) support, and integrated hardware power-loss protection.
At its core, the PD5 uses a Silicon Motion SM8366 16-channel PCIe 5.0 x4 controller paired with 232-layer 3D TLC NAND flash. A standout technology is Exascend’s Dual Power Loss Protection, which uses custom firmware and onboard tantalum capacitors to flush DRAM cache data to NAND during unexpected power failures, then rapidly reconstructs the Flash Translation Layer mapping table upon power restoration. The drive is offered in standard and Pro Editions with different endurance ratings, carries a 5-year limited warranty, and has active power consumption up to 25W with 5.4W idle.
Samsung 990 QLC DRAMless Gen4 2TB SSD Review Choosing Between This the TLC-based 990 EVO Plus
2 d ago
Longsys has launched its Gen5 2TB mSSD (Micro SSD), a single-chip System-in-Package (SiP) drive capable of exceeding 12GB/s transfer speeds and scaling to 8TB, targeting AI PCs and storage-constrained devices. The drive, received directly from Longsys Chairman Huabo Cai, represents a major leap from the company’s Gen4 mSSD, which delivered 7.1GB/s from a quarter-sized package. The Gen5 mSSD integrates the entire storage solution into one SiP, eliminating nearly 1,000 solder joints compared to conventional SSDs, cutting manufacturing defect rates from ~1,000 DPPM to just 100, and reducing production costs by roughly 10 percent. At its heart is the Maxio MAP1802, a DRAMless PCIe Gen5 x4 NVMe controller built on a 6nm process, supporting NAND interface speeds of 4,800MT/s. The sample arrived in a gold aluminum heatsink with a copper vapor chamber for efficient thermal management. Available in the compact 2230 form factor (22mm × 30mm), the mSSD is suitable for handheld gaming systems and ultrathin laptops. While Longsys has not confirmed the NAND flash, industry speculation points to YMTC’s X4-9070 3D NAND based on Xtacking 4.0 technology, as achieving 8TB and 12GB/s-class performance would require the latest high-density flash. The Gen5 mSSD is now reaching monthly production capacity of over 1 million units, directly supplying AI PCs and optimized for KV Cache workloads. The evolution from 2019’s Toshiba BG4 (1TB SiP at 2GB/s) to today’s 13GB/s-capable Gen5 mSSD highlights rapid SiP SSD advancement.
**SEO News Summary: MRDIMM Gen2 and DDR5-8000 RDIMMs Target Next-Gen Server Memory Bandwidth**
For readers tracking server memory, AI chip, and data center infrastructure trends, the core news from Computex is that next-generation x86 server platforms from Intel and AMD are driving rapid memory advancements, with Micron launching DDR5-8000 RDIMMs and Samsung developing MRDIMM Gen2 reaching 12,800 MT/s. Micron’s DDR5-8000 RDIMMs, showcased at the event, double current DDR5-6400 speeds and support high-capacity 256GB modules using 32Gbit dies on its 1y process, offering better energy efficiency per picojoule. Meanwhile, Samsung’s MRDIMM Gen2 pushes multiplexer rank DIMMs to 12,800 MT/s—a 45% bandwidth increase over Gen1’s 8,800 MT/s. Using DDR5-6400 memory in interleaved mode, the new MRDIMMs top out at 128GB capacity, trading off capacity for speed versus standard RDIMMs. Samsung targets AMD’s EPYC “Venice” platform (expected later this year with 16 memory channels) for MRDIMM Gen2, potentially delivering 1.6 TB/s total bandwidth—a massive jump from today’s 0.6 TB/s on 12-channel DDR5-6400 EPYC 9005. Intel’s Diamond Rapids Xeon (2027) also supports 16 channels. These developments underscore that higher clocked RDIMMs and MRDIMMs, alongside additional memory channels, are critical to alleviating memory bandwidth bottlenecks in next-gen servers and data centers.
On a recent podcast, DIGITIMES analyst Luke Lin reviewed two major updates in the chip industry: Intel's Ireland fab expansion and rising US scrutiny of Samsung Electronics and SK Hynix as both South Korean chipmakers expand memory capacity at home. He also...
The 3rd edition of the RAISE Summit 2026 in Paris attracted approximately 9,000 attendees and showcased major AI trends including sovereignty, inference, and the rise of neoclouds, though the venue proved too small for the event's scale. Organizers reported strong participation from global exhibitors and high-profile participants, but the prestigious central Paris location became overcrowded. Conference sessions limited to 20 minutes felt rushed. AI dominated the summit, but exhibition layout was disjointed, mixing infrastructure vendors with financial services. Notable absent exhibitors included Dell, HPE, Oracle, Intel, IBM, and Anthropic, while storage and SSD vendors were scarce. Neocloud players like CoreWeave and Crusoe were present, and new AI hardware vendors such as Cerebras and SambaNova took the stage. European tech presence remained limited despite sovereignty discussions. Key trends highlighted included investment dynamics, open-source AI, AI coding tools, token consumption optimization, and energy efficiency. The RAISE Summit 2027 will move to the larger Palais des Congrès, Porte Maillot, Paris, from July 6–8, 2027.
Apple Swallowed SigLens from SigScalr Wishing to improve its applications monitoring By Philippe Nicolas | July 17, 2026 at 2:01 pm Apple has acquired select assets from SigScalr, the developer behind the application monitoring platform SigLens, according to a European Union filing.The transaction, reported to the EU on 12 March 2026, also includes an option for Apple to hire certain SigScalr employees. As usual with Apple, the web site of the prey has been studown. SigScalr developed SigLens, an open-source observability platform that enables organizations to monitor and analyze logs and application activity across distributed systems. The company claimed the platform delivered up to 100% greater efficiency than competing offerings from Datadog and Splunk. Founded in May 2021 by Kunal Nawale, SigScalr was headquartered in Nashua, New Hampshire, with additional offices in California, Montana, and India. According to its LinkedIn profile, the startup employed between two and ten people. In 2024, SigScalr raised $1.76 million in pre-seed funding and later open-sourced SigLens in February of that year.
Chinese DRAM maker ChangXin Memory Technology's (CXMT) IPO has entered its final stage, drawing close market attention. PC supply chain sources said the US had briefly planned to add CXMT to its entity list but had not announced it, and that easing US-China...
Druva, a provider of data security, announced it has been recognized as a Leader in the 2026 Gartner Magic Quadrant for Backup and Data Protection Platforms.This marks Druva’s second consecutive year as a Leader. Enterprise data protection is entering a new phase as AI becomes embedded across business workflows and identity becomes increasingly connected to how data is accessed, governed and recovered. Backup and recovery can no longer operate as isolated infrastructure. When something goes wrong, IT and security teams need to understand what changed, determine what can be trusted, and restore operations with confidence. That requires a context-aware foundation connecting data, metadata, identities, permissions and recovery history. Druva was built with a fully managed SaaS architecture that isolates backup data and credentials from the primary network while reducing customer-managed infrastructure. Through Dru MetaGraph, Druva brings graph-based context to backup and recovery, helping organizations map relationships across data, metadata, identity and permissions. With DruAI, the platform extends this foundation with AI-driven investigation, identity-aware recovery, and agentic workflows. Customers can restore not only the data, but the identity and permission context around it without managing backup infrastructure. “AI is accelerating both cyber threats and day-to-day change across the enterprise,” said Jaspreet Singh, CEO, Druva. “As AI becomes part of more workflows, data protection has to preserve the context organizations need when something goes wrong, including what changed, which identities and permissions were involved, and which data can be trusted. In our opinion, being named a Leader for the second consecutive year reflects the growing need for fully managed SaaS data protection built for this level of speed, context and security.”
A rendering of SK Hynix’s Yongin Y1 fab, where the company is preparing equipment installation for the first cleanroom phase. Credit: SK Hynix SK Hynix has begun placing orders with major suppliers for advanced DRAM manufacturing equipment for the first phase of its Yongin Y1 fab, with the initial installation expected to support production of about 20,000 wafers per month, according to ZDNet...
Lenovo has launched the ThinkStation P3 Ultra SFF Gen 2, a mini-PC workstation that bridges the gap between compact desktops and small towers, targeting users seeking AI-capable, expandable workstation hardware. Billed as an AI workstation, the 3.9-liter system combines Intel’s Core Ultra 9 285 processor (8P+16E, up to 5.6GHz) with an optional NVIDIA RTX 4000 Ada Generation SFF GPU (20GB GDDR6) and up to 64GB DDR5-6400 memory. Storage is handled by a 1TB PCIe Gen5 x4 SSD.
The SFF form factor enables two low-profile PCIe slots (one x16 Gen4, one x8 Gen4) for expansion, distinguishing it from Lenovo’s smaller 1-liter P3 Tiny Gen 2. The chassis emphasizes airflow with front grating and supports both horizontal and vertical orientations. Front I/O includes two USB-C ports at 20Gbps (USB4) and one USB-A 10Gbps; the rear offers four USB-A 10Gbps, three DisplayPort 1.4 outputs (wired to integrated Intel graphics), and a single gigabit Ethernet port—notably lacking 2.5GbE or HDMI, though Lenovo offers an optional Thunderbolt 4 upgrade. The NVIDIA card provides four mini-DisplayPort 1.4 outputs. Connectivity includes Wi-Fi 7 and Bluetooth 5.4 via Intel BE200, powered by a 330W external PSU.
Measuring 202 x 87 x 223 mm and weighing 1.8 kg, the P3 Ultra SFF Gen 2 targets professionals needing a compact yet upgradable AI workstation. The system’s design leverages extra space for dual-slot GPUs and multiple PCIe cards, making it a larger, more capable alternative to Lenovo’s P3 Tiny line.
When it comes to SSDs, I can’t help but wonder whether the latest price spike is simply a temporary shock driven by AI demand, or whether it was inevitable all along. For a while, we had it incredibly good. High-performance, DRAM-equipped SSDs with generous capacities and five-year warranties were selling at prices that would seem unbelievable today. Those days are clearly behind us, and I have a hard time believing pricing will ever return to those levels. We really did have it good. What has become increasingly obvious is that the storage industry needed to evolve, and two technologies have driven that change more than any others: QLC NAND and the widespread adoption of Host Memory Buffer (HMB) SSDs. Don’t get me wrong. Both technologies had their share of growing pains. Early implementations weren’t always impressive, and skepticism was justified. But today they’ve become mainstream for one simple reason: the average SSD user simply can’t tell the difference between a modern QLC, HMB-based SSD and a premium DRAM-equipped model. That’s just a fact. In truth, the typical user can’t distinguish between an entry-level SSD and the fastest consumer SSD on the market. I’d even argue that most experts couldn’t reliably tell the difference in normal, everyday use without benchmarks. That’s where Samsung’s Gen4 QLC DRAM-less SSD enters the picture. It isn’t trying to compete by posting record-breaking benchmark numbers. It’s designed around the reality of how most people actually use their PCs. The Samsung 990 is something of a misnomer, in my opinion. Historically, SSD product lines have followed a predictable path: manufacturers introduce a standard model first, then follow it with a higher-performance “Pro” version. Samsung took a different approach with the 990 series. The Samsung 990 Pro debuted in October 2022 (reviewed here), followed by the more economical 990 EVO, a TLC-based DRAMless SSD (reviewed here). In 2024, Samsung raised the bar again with the 990 EVO Plus, a faster TLC DRAMless drive (reviewed here) that quickly earned an excellent reputation for delivering strong performance without the added cost of onboard DRAM. Traditionally, Pro-series SSDs were designed, and priced, for enthusiasts, gamers, and professional users, while the standard model targeted everyday consumers. This time, however, Samsung appears to have reversed course. The new 990 Gen4 SSD doesn’t quite match the overall performance of its Pro sibling, but it seems intended to strike a more attractive balance between speed, efficiency, and value for mainstream buyers. The one disappointing compromise is the warranty. Samsung has reduced coverage to a three-year limited warranty, stepping back from the five-year warranty that has become the industry standard for quality consumer SSDs. The Samsung 990 is a PCIe 4.0 (Gen4) x4, four-channel M.2 2280 SSD (22mm x 80mm) that supports the latest NVMe 2.0 protocol. At launch, it is available in 1TB and 2TB capacities, delivering sequential performance of up to 7,250MB/s read and 6,450MB/s write, along with random performance of up to 1.2 million IOPS. The drive features a single-sided, DRAMless design, relying instead on your system’s Host Memory Buffer (HMB) rather than dedicated onboard DRAM. This approach reduces manufacturing costs while still delivering performance that should satisfy the vast majority of everyday users. All Samsung 990-series SSDs have been built around variations of what we now recognize as Samsung’s “Piccolo” controller. Looking closely at this latest version, however, you’ll notice the controller is marked “PiccoloQ,” with the “Q” almost certainly denoting its use with QLC NAND flash memory. This would lead one to believe the NAND of choice is most likely Samsung’s 9th-gen 280-layer QLC V-NAND. This controller has the number designation of S4LY028. Examining our 2TB sample, we find a single Samsung V-NAND QLC flash package on the PCB, with an empty footprint directly beside it. That unused space certainly raises the possibility that Samsung may eventually introduce a higher-capacity 4TB model using a second NAND package. Complete SSD specs can be seen in the Samsung release republished here: There is one thing that is perhaps a highlight of this SSD. It is single-sided. With only a total of two chips on the 990, we simply have to believe that this SSD is a very cool running SSD, especially when those listed specifications only list a high of 4.3 watt active power consumption. The Samsung 990 Gen4 SSD comes with a 3-year limited warranty and perhaps one of the lowest TBW ratings we have seen for a while which is 400TBW for the 1TB and double that for the 2TB capacity. Don’t worry though. The typical user still won’t ever reach ‘end life’ with the 990. MSRP pricing is listed at $269.99 (1TB) and $529.99 (2TB) and availability will be present at Amazon and other e-tailers as of this report release.
**ISC Conference 2025 News: HPC, AI, and Quantum Converage Drives Storage and Infrastructure Shifts**
The ISC conference in Hamburg, Germany, highlighted the convergence of HPC, AI, and quantum computing as core themes, with LineShine from Shenzhen, China, claiming the top spot on the TOP500 list. Key topics included AI at scale inference, energy-efficient cooling, and data infrastructure, with storage vendors riding renewed interest in tiering due to steep SSD pricing, reinforcing HDD relevance. KV cache storage was heavily promoted for efficient inference, which now overshadows training in conversations. Edge AI also had a presence. Notable exhibitors like Seagate, Toshiba, WD, Kioxia, Samsung, Micron, and Dapustor showcased solutions, while data management vendors such as Arcitecta, Atempo, DDN, and Quantum promoted global data access and compute-in-place amid data volume challenges. File storage (BeeGFS, Vast Data, Weka) and object storage (MinIO, Spectra Logic) systems, including Lustre and DAOS, were prominent. Sovereignty remained a key European topic. Notable absences included booths from Astera Labs, Broadcom, Intel, SK hynix, and NetApp. Looking ahead, the next ISC is scheduled for June 7–11, 2027, in Hamburg, as the event continues to evolve from its HPC roots.
Apple has announced a new multiyear commitment with Broadcom to design and produce custom silicon and advanced wireless connectivity components, with the deal expected to exceed $30 billion. The partnership, Apple’s largest under its American Manufacturing Program (AMP), will generate over 15 billion US-made chips and support hundreds of American jobs. Under the agreement, Broadcom will invest $1.5 billion to expand and modernize its Fort Collins, Colorado facilities — producing advanced radio frequency components including FBAR filters and cutting-edge wireless technologies for Apple products. The announcement advances Apple’s goal of building an end-to-end silicon supply chain in the US, building on its broader $600 billion US economy investment over four years. Apple CEO Tim Cook stated the partnership accelerates American manufacturing innovation. Broadcom CEO Hock Tan highlighted the expanded Fort Collins footprint as a hub for groundbreaking connectivity tech. Key entities: Apple, Broadcom, Fort Collins. Primary keywords: custom silicon, wireless connectivity, chip supply chain, US manufacturing, AMP. Search intent matches semiconductor and chip supply chain news.
Read also : OpenAI Partners with Cerebras OpenAI is partnering with Cerebras to add 750MW of ultra low-latency AI compute to our platform Top News | By Philippe Nicolas | January 16, 2026 | News
Cerebras Systems, maker of a fast AI infrastructure, announced a major expansion of its European infrastructure footprint.Cerebras will bring its first European data center capacity online by the end of 2026, with rapid build-out across France and the Nordics. The company plans to expand total capacity to 200 MW by the end of 2027, with a portion of that capacity expected to support OpenAI workloads as part of the companies’ existing partnership. The expansion will bring Cerebras’ high-speed AI inference infrastructure closer to European users, helping deliver faster response times for increasingly complex AI workloads. “We are contracting significant capacity for 2027, with data centers slated for Norway and Finland as we actively build across Europe,” said Andrew Feldman, co-founder and CEO, Cerebras. “These deployments will enable us to move decisively on what our customers have been asking for: fast, high-performance AI compute located in Europe.” Frontier compute for Europe As AI models support increasingly complex and interactive workloads, demand for local, low-latency AI infrastructure has surged across European enterprises, research institutions, and governments seeking alternatives to compute capacity concentrated in the US and Asia. Cerebras’ wafer-scale architecture is designed to deliver industry-leading inference and training performance, and the company’s European build-out positions it to serve this demand directly from within the region. “Our customers don’t just want AI compute. They want it close to home, powered responsibly, and available fast,” added Feldman. “This expansion and capacity plan reflects our confidence in Europe as a long-term growth market for Cerebras.” Cerebras at RAISE Summit Cerebras co-founder and CEO Andrew Feldman participated to RAISE Summit in Paris.