Graph neural networks (GNNs), which have emerged as an effective method for handling machine learning tasks on graphs, bring a new approach to building recommender systems, where the task of recommendation can be formulated as the link prediction problem on user-item bipartite graphs. Training GNN-based recommender systems (GNNRecSys) on large graphs incurs a large memory footprint, easily exceeding the DRAM capacity on a typical server. Existing solutions resort to distributed subgraph training, which is inefficient due to the high cost of dynamically constructing subgraphs and significant redundancy across subgraphs. The emerging persistent memory technologies provide a significantly larger memory capacity than DRAMs at an affordable cost, making single-machine GNNRecSys training feasible, which eliminates the inefficiencies in distributed training. One major concern of using persistent memory devices for GNNRecSys is their relatively low bandwidth compared with DRAMs. This limitation can be particularly detrimental to achieving high performance for GNNRecSys workloads since their dominant compute kernels are sparse and memory access intensive. To understand whether persistent memory is a good fit for GNNRecSys training, we perform an in-depth characterization of GNNRecSys workloads and a comprehensive analysis of their performance on a persistent memory device, namely, Intel Optane. Based on the analysis, we provide guidance on how to configure Optane for GNNRecSys workloads. Furthermore, we present techniques for large-batch training to fully realize the advantages of single-machine GNNRecSys training. Our experiment results show that with the tuned batch size and optimal system configuration, Optane-based single-machine GNNRecSys training outperforms distributed training by a large margin, especially when handling deep GNN models.

If you've ever waited for a webpage to load, took a lap around the office while your computer rendered 4K video or watched in dismay as your favorite video game stuttered to life, you know that computers are struggling to keep up with all the data. On Monday, the chip maker announced the ship date for Optane, a memory module that can change the data game. It takes the speed and fluidity of RAM and combines it with the storage capabilities of Flash memory. SEE ALSO: Trekkies can now wake up their Amazon Echo by saying'computer' "Storage needs dramatically increased over the last several years," said Intel Client Computing Group SVP Navin Shenoy who caught up with me to explain why Optane (unveiled at CES 2017) is a potential game-changer for computer users. Working with system software, Optane can pre-cache frequently-used apps, which means they could load almost instantly.

Though some critics love to knock PCs as dinosaurs, laptops and desktops have gotten sexier, faster and even smarter. For every blue screen of death, there are droves of technological enhancements driving PCs into the era of virtual reality, 4K video and 5G connectivity. Here are the top 10 PC technology and trends to watch next year. An Intel employee demonstrates the company's Project Alloy headset on stage during IDF 2016 in San Francisco on August 16, 2016. VR devices will come in many new shapes and sizes, with some of them acting essentially as PCs that fit on your head.