As bytes-per-FLOP ratios continue to decline, communication is becoming a bottleneck for
performance scaling. This paper describes bandwidth steering in HPC using emerging …

Recent advances in integrated photonics enable the implementation of reconfigurable, high-
bandwidth, and low energy-per-bit interconnects in next-generation data centers. We …

We demonstrate a disaggregated optically-connected memory (OCM) architecture that
supports dynamic resource allocation. Our proposed architecture utilizes a hitless …

The influx of new data-intensive applications, such as machine learning and artificial
intelligence, in high performance computing (HPC) and data centers (DC), has driven the …

We investigate the performance of cloud network with optically disaggregated memory
under different CPU-memory interconnection architectures. We found that long burst of …

Provided are a memory access method and a server for performing the same. A memory
access method performed by an optical interleaver included in a server includes receiving a …

Driven by the substantial resource under-utilization encountered in modern mega-DCs, that
can reach up to 50%[2.1]-[2.2], significant research efforts have focused in the direction of …

As bytes-per-FLOP ratios continue to decline, communication is becoming a bottleneck for
performance scaling. This paper describes bandwidth steering in HPC using emerging …