Look Up Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Say More Safety Net Creator Hub Versus Gift Ideas For Everyone On Your List Mashable Selects Switch Off Trending Now In My Bag VidCon with Mashable All Series Best Buy's 60th Anniversary Sale is live: Shop the best deals on TVs, laptops, Apple gear If you missed Prime Day, these are some of the next-best discounts of the summer. Christina Buff is a Nashville-based freelance writer for who covers shopping with a splash of entertainment. If you're ever wondering what streaming service you need to watch something (and the cheapest way to sign up for it), she's your girl. All products featured here are independently selected by our editors and writers. If you buy something through links on our site, Mashable may earn an affiliate commission.

Say More Look Up Safety Net Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Creator Hub Versus Gift Ideas For Everyone On Your List Mashable Selects Switch Off Trending Now In My Bag VidCon with Mashable All Series Shop SSD deals at Amazon, Best Buy and more to save up to $300. Samantha Mangino, is an e-reader expert. She's tested all the Kindle and Kobo e-readers, diving into the buzziest models. Her love of reading extends beyond the e-reader beat, covering books and authors, too. In the tech space, she covers tablets, smart rings, and headphones.

Gift Ideas For Everyone On Your List Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Look Up Mashable Selects Say More Versus Creator Hub Switch Off Trending Now Safety Net In My Bag VidCon with Mashable All Series If you're tired of deleting old games to make room for new ones, this 2TB SSD is on sale for $279 right now. Tabitha Britt is an award-winning freelance journalist, editor, and SEO/AEO strategist. Aside from reviewing dating apps and sex toys for Mashable, Tabitha is also the founding editor-in-chief of DO YOU ENDO -- a digital magazine by individuals with endometriosis, for individuals with endometriosis. She has a Master's degree in Creative Publishing and Critical Journalism from The New School for Social Research and is a grad of Sextech School. You can find more of her work in various online publications, including,, and .

Look Up Say More Versus Creator Hub Switch Off Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Trending Now Safety Net In My Bag VidCon with Mashable Back to School Furtastic All Series Both internal and external SSDs are on sale. Samantha Mangino, is an e-reader expert. She's tested all the Kindle and Kobo e-readers, diving into the buzziest models. Her love of reading extends beyond the e-reader beat, covering books and authors, too. In the tech space, she covers tablets, smart rings, and headphones.

Look Up Say More Versus Creator Hub Switch Off Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Trending Now Safety Net In My Bag VidCon with Mashable Back to School Furtastic All Series Here's how the MacBook Neo compares to older MacBook Airs Haley Henschel is a Chicago-based Senior Shopping Reporter at Mashable who reviews and finds deals on popular tech, from laptops to gaming consoles and VPNs. She has years of experience covering shopping holidays and can tell you what's actually worth buying on Black Friday and Amazon Prime Day. Her work has also explored the driving forces behind digital trends within the shopping sphere, from dupes to 12-foot skeletons . All products featured here are independently selected by our editors and writers. If you buy something through links on our site, Mashable may earn an affiliate commission.

Look Up Say More Versus Creator Hub Switch Off Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Trending Now Safety Net In My Bag VidCon with Mashable Back to School Furtastic All Series Lauren Allain is a freelance journalist covering deals at Mashable. She graduated from Western Washington University with a B.A. in journalism and holds an M.B.A from Webster Leiden. You can find more of her work online from publications including Reader's Digest, U.S. News & World Report, Seattle Refined, and more. When she's not writing, Lauren prefers to be outside hiking, bouldering, swimming, or searching for the perfect location for all three. All products featured here are independently selected by our editors and writers.

When you purchase through links in our articles, we may earn a small commission. Samsung's brand-new 990 SSD just hit its lowest price yet: $370 for 2TB Samsung's newest SSD with up to 7,250 (read) MB/s and 6,450 (write) MB/s just got its first price cut: 2TB for $370, down from its $530 MSRP. It's a rough time for anyone who's looking to upgrade their PC hardware. The AI-driven shortage has spiked the prices of memory and storage, doubling or even tripling in some cases. If you need better and/or more internal storage and can't wait around for prices to fall back down (if they ever do), here's a deal worth jumping on. Samsung released its new baseline 990 SSD last month with an MSRP of $530.

Large-language-model inference with long contexts often produces key-value (KV) caches whose footprint exceeds the capacity of high-bandwidth memory on a GPU. Prior LLM inference frameworks such as vLLM mitigate this pressure by swapping KV cache pages to host DRAM. However, the high cost of large DRAM pools makes this solution economically unattractive. Although offloading to SSDs can be a cost-effective way to expand memory capacity relative to DRAM, conventional frameworks such as FlexGen experience a substantial throughput drop since the data path that routes SSD traffic through CPU to GPU is severely bandwidth-constrained. To overcome these limitations, we introduce HiFC, a novel DRAM-free swapping scheme that enables direct access to SSD-resident memory with low latency and high effective bandwidth. HiFC stores KV pages in pseudoSLC (pSLC) regions of commodity NVMe SSDs, sustaining high throughput under sequential I/O and improving write endurance by up to 8 . Leveraging GPU Direct Storage, HiFC enables direct transfers between SSD and GPU, bypassing host DRAM and alleviating PCIe bottlenecks. HiFC employs fine-grained block mapping to confine writes to high-performance pSLC zones, stabilizing latency and throughput under load. HiFC achieves inference throughput comparable to DRAMbased swapping under diverse long-context workloads, such as NarrativeQA, while significantly lowering the memory expansion cost of a GPU server system by 4.5 over three years.

When you purchase through links in our articles, we may earn a small commission. This good-performing single NVMe, USB4 drive enclosure makes it supremely easy to install an SSD, and looks nice in the process. Asus's PA40SU is a good-looking 40Gbps enclosure with on-par performance and exceptionally easy, tool-less SSD installation. Our only caveat is that it runs quite warm despite an internal fan. But despite its built-in fan, it can still get uncomfortably warm under sustained transfers.

We present the design and implementation of a new lifetime-aware tensor offloading framework for GPU memory expansion using low-cost PCIe-based solid-state drives (SSDs). Our framework, TERAIO, is developed explicitly for large language model (LLM) training with multiple GPUs and multiple SSDs. Its design is driven by our observation that the active tensors take only a small fraction (1.7% on average) of allocated GPU memory in each LLM training iteration, the inactive tensors are usually large and will not be used for a long period of time, creating ample opportunities for offloading/prefetching tensors to/from slow SSDs without stalling the GPU training process. TERAIO accurately estimates the lifetime (active period of time in GPU memory) of each tensor with the profiling of the first few iterations in the training process. With the tensor lifetime analysis, TERAIO will generate an optimized tensor offloading/prefetching plan and integrate it into the compiled LLM program via PyTorch. TERAIO has a runtime tensor migration engine to execute the offloading/prefetching plan via GPUDirect storage, which allows direct tensor migration between GPUs and SSDs for alleviating the CPU bottleneck and maximizing the SSD bandwidth utilization. In comparison with state-of-the-art studies such as ZeRO-Offload and ZeRO-Infinity, we show that TERAIO improves the training performance of various LLMs by 1.47 on average, and achieves 80.7% of the ideal performance assuming unlimited GPU memory.