Since 1992 Boston's mantra has been to excel in all our activities. With a dedication to helping our partners succeed, we strive to always provide our clients with first-to-market technology and award-winning, tailored solutions that are customised to fit their exact specifications
Inspiring new generations of innovation and advancements in technology, Boston were delighted to unveil many of our much-awaited storage and sever solutions to attendees of SC17.
To refresh your memory there is an overview below, with links to more information and an enquiry form - we'd love to hear from you!
Boston Flash-IO Talyn - All-flash NVMe over Fabrics
The Boston Flash-IO Talyn was by far the most popular solution on the stand at this years Supercomputing. Perhaps it was the fact that this is a competition-busting piece of technology or it's stylish bright blue bezel and attractive vinyl wrap...either way there was a steady stream of interest in this solution that features award-winning NVMesh server SAN.
Developed in partnership with Excelero, Mellanox, Micron and Supermicro, the Boston Flash-IO Talyn addresses the pressure placed on storage architectures by data-heavy applications across a range of verticals such as HPC, rendering, Media & Entertainment, deep learning and AI.
Boston ANNA Volta - Featuring 8x NVIDIA® Volta™ GPUs and taking advantage of next generation NVLink™
The Boston ANNA Volta XL, features 8x NVIDIA® Tesla® V100 GPU accelerators for deep learning training, purpose-built to dramatically reduce training time. Running Caffe and Torch on the Tesla V100 delivers the same model within days versus weeks on CPU based compute systems.
Boston Quattro 12256-T "LHP Edition" - Liquid cooled Intel® Xeon Phi™ 2U, 4 Node server
The Boston Quattro 12256-T “LHP edition” is the 2U, 4-node server featuring the latest generation Intel® Xeon Phi processor (codenamed Knights Landing) is designed with parallelism and vectorised workloads in mind. Intel’s 2nd generation Intel® Xeon Phi Product Family, delivers massive thread parallelism, data parallelism and memory bandwidth.
In collaboration with Calyos, Boston have now developed an improved version of this server SKU by utilising their cooling solution based on LHP (Loop Heat Pipe) technology. This innovative 2-Phase cooling solution has been designed for the datacentre and provides several advantages such as improved thermals, better efficiency, lower power consumption and extended server operating temperature when compared to a standard air-cooled equivalent.
Half-Width 16-port Non-blocking 100GbE Open Ethernet Switch System. The SN2100 switch provides the highest density of 25G ports per 1RU with 128 ports in a side-by-side configuration to provide highest density as well as ToR redundancy in the smallest footprint.
Boston dataScaler-L SFF - Intel® Lustre* Small Form Factor Solution
Boston’s dataScaler-L SFF (small form factor) appliance is the latest addition to Boston’s Lustre based parallel file system portfolio and provides a lower cost / capacity entry point for Lustre storage systems. Traditional Lustre solutions on the market are focused on addressing the high end cluster users, while the Boston Lustre SFF is designed for small or mid-sized clusters with capacities as low as 18TB (and scaling to 720TB). This solution provides a high performance alternative to standard NFS and provides customers with the option to scale and grow as the cluster and user demands increase over time.
Boston Limited's latest 2U Quattro server, the Boston Quattro 22128-T, features the latest generation Intel® Xeon® Scalable Family processors (formerly codenamed Skylake). The Boston Quattro 22128-T is designed for scalable and reliable performance, with the greatest variety of features and integrations designed to meet the needs of the widest variety of workloads.
Intel's 6th generation Intel® Xeon® Scalable Family, delivers 1.5X greater memory bandwidth with 6 memory channels and up to 2TB capacity.
The new Intel® Xeon® Scalable Family also provides up to 28-cores per processor with up to 2 threads per core, giving a total of 56 threads per processor which are, of course, binary compatible with software compiled for existing x86 hardware.
Coupled with optional single on-package Intel® Omni-Path interface incremental to existing 48 PCIe lanes and single cable connection to QFSP I/O module, the Boston Quattro 22128-T offers a truly integrated, high performance, low latency compute platform.
Of course, being part of the Boston Quattro range, the Boston Quattro 22128-T is of a multi-node design offering 4 independent compute nodes (all based on the same Intel® Xeon® Scalable Family technology) each with multiple HDD/SSD support, independent platform management, redundant and optimised cooling and titanium-level redundant power supplies.
To help our clients make informed decisions about new technologies, we have opened up our research & development facilities and actively encourage customers to try the latest platforms using their own tools and if necessary together with their existing hardware. Remote access is also available