Nvidia claims to offer "the world's most powerful AI system" in a more compact and energy efficient package with the DGX-2-- a turnkey HPC system promising to deliver up to 2 petaFLOPs of computing power.

As a followup to the DGX-1, the DGX-2 carries x16 fully interconnected Tesla V100 32GB GPUs (double the amount of the 2017 DGX-1). It also includes up to 1.5TB of system memory, 30/60TB of NVMe storage, InfiniBand or 100GBe networking and x2 Xeon Platinum processors. Tying the GPUs together is what Nvidia calls "NVSwitch," an interconnect fabric allowing the GPUs to communicate with each other in the system at 300GB/s, making 14TB/s of aggregate system bandwidth.

ATEN Technology expands its enterprise KVM solution line with a line of Secure KVM switches-- 18 products all compliant with PSS PP v3.0 (Protection Profile for Peripheral Sharing Switch, Version 3.0).

The PSS P v3.0 Secure KVM Switches isolate computer sources and peripherals while sharing a single keyboard, mouse, monitor, speaker set and Common Access Card (CAC) reader between connected computers with various security classifications. Compliance with PSS PP v3.0 ensures peripheral sharing capabilities provide maximum user data security when switching port focus, preventing unauthorised data flows or leakage between connected sources.

Semiconductor specialist Xilinx presents what it claims is a chip with capabilities superseding the field-programmable gate array (FPGA)-- the adaptive compute acceleration platform (APAC).

FPGAs are integrated circuits one can further configured after manufacturing. They find use as accelerators in large-scale high performance computing (HPC) and supercomputing environments, as well as cloud computing datacentres and Internet of Things (IoT) hardware. Xilinx says the ACAP goes a step beyond FPGAs. A "highly integrated multi-core heterogeneous compute platform," the ACAP can not only be changed at the hardware level to adapt to a range of applications and workloads, but adapts dynamically during operations for "levels of performance and performance per-watt that is unmatched by CPUs or GPUs."

According to Gartner, global integrated systems revenue is to reach $12.3 billion in 2018-- an 18.4% increase over the $10.2bn of 2017, with hyperconverged integrated systems (HCIS) seeing the strongest growth.

Integrated systems combine server, storage and network infrastructure with management software handling the provisioning and management of the combined unit. It derives the benefit of an architectured design and deployment of integrated compute, storage and memory infrastructure.

"The majority of integrated systems replace existing infrastructure, which is great for cost, agility and consolidation of IT and efficiency metrics," the analyst says. "When implementing this as part of a digital business initiative, however, IT organisations must look at how the potential savings of capital expenditure (capex), may be offset by potential shifts in operating expenditure (opex)."