We de-lid an Intel Optane DC Persistent Memory module to see what chips go into making one of 2019's most anticipated technologies. In the process, we found a few surprises and new clues as to how these work
Packet is putting a different spin on microservers. The company's early 2019 microserver offering will utilize Netronome SmartNICs along with Ampere eMAG Arm CPUs, Intel CPUs, and what the company may have inadvertently tipped off as AMD EPYC 3000 series CPUs during our discussion
At Intel Architecture Day 2018, the company showed off a slide showing its platforms. These platforms included an uptick in FPGA boards as well as the Intel Xeon D series.
This week, Google announced a low-cost and fast deep learning model training in GCP. With the results of MLPerf v0.5 in hand and the Google TPU v2 Pod hardware in its cloud, Google showed off the fast and cost-effective training solution
Our Intel Optane 905P 380GB M.2 NVMe SSD review shows why this is the best solution for write-heavy workloads in servers with an empty M.2 22110 slot.
We show why Intel Foveros is awesome as a new 3D chip stacking technology set to disrupt EMIB as the go-to multi-chip packaging option for Intel
New 1U and 2U Marvell ThunderX2 servers are available with built-in SAS controllers and 10GbE networking paired with dual 32 core 128 thread Arm CPUs
At a launch event today, the NVIDIA Jetson AGX Xavier module is now available for robotics applications shipping in volume, not just for developer kits. NVIDIA also announced a new NVIDIA Jetson TX2 4GB module for lower cost embedded applications
At Intel Architecture Day 2018, we got fresh Intel Sunny Cove microarchitecture details which show why Intel is about to make a much larger jump in per clock performance than we have seen in many generations
NVIDIA DGX-2H chassis chained together in a 32x system set along with clusters of DGX-1 and DGX-2 systems set six world record benchmarks on the initial set of MLPerf v0.5 results launched today