STAC-A2 (derivatives risk) on 8 x Ampere GPUs in an NVIDIA DGX A100
kdb+ 3.6 with Dell EMC PowerScale F200 under STAC-M3 (tick analytics)
Arista DCS-7130-48L under STAC-TS (time sync and capture)
New LDA/Xilinx solution under STAC-T0 (tick-to-trade network I/O)
First STAC-M3 results from new generations of CPU and Optane Persistent Memory
NVIDIA DGX A100 system with new GPUs sets three records
NFS-based solution with inline compression sets storage efficiency record
Solution sets multiple throughput records
First solution with distributed Optane Persistent Memory sets 8 records across scale and baseline benchmarks.
For comparison to previous bare metal results on the same hardware
Thanks to all of you who made Global STAC Live Fall 2020 a great success. While we missed seeing you in person this year, the STAC community made the most of remote events. In fact, our full-year global attendance was up by more than 25% over 2019.
This fall’s Global STAC Live covered important topics, including: evolving technologies and architectures for time series analytics; distributing, processing, and visualizing realtime data; the requirements for moving exchanges to the cloud and the progress already made; maturing the development and deployment of custom FPGA-based solutions; and the impact of new technologies on ML/AI workflows, including how to choose the right platforms to facilitate your AI projects.
The agenda, recorded presentations, and slides from the events are available here.
Plans are now underway for the Spring 2021 STAC Summits. <more>
The financial technology community has one responsibility in the COVID-19 crisis that only it can fulfill.
The STAC-TS Working Group has poured a lot of effort into developing standards and software for demonstrating timestamp accuracy to the satisfaction of European regulators concerned with RTS 25 of MiFID 2. Much of that effort has concerned software-application timestamps. App timestamping is central to many firm's plans but is one of the most misunderstood issues with RTS 25.