Installed in late 2019, Siku (meaning “sea ice” in Inuktitut) was funded, in large part, by the Atlantic Canada Opportunities Agency (ACOA) with the intention of generating regional economic benefits through industry engagement, while recognizing the important work that ACENET does for academic research in the region. The system is to be used in a manner which leads it to be financially self-sustaining.
Siku gives preferential access to paid industry usage, but gladly offers surplus capacity to academic researchers, at no cost. The pay-for-use model serving regional industry is also available to academic researchers seeking comparable access to Siku.
Located at Memorial University, Siku is a 2300 core computing cluster dedicated to industry engagement and advancing regional research priorities. Siku incorporates Intel Cascade Lake CPUs, a high-throughput, low-latency EDR Infiniband interconnect, AI-capable NVIDIA Tesla V100 GPUs, a 1.5 PB parallel filesystem, tape back-up, and offers both batch and cloud-computing interfaces.
Siku is not intended to divert academic researchers from the national infrastructure, but rather to address needs and priorities that cannot be readily accommodated by Béluga, Graham, Niagara, Cedar or Arbutus. To find out whether Siku is the right fit for your academic research, submit the light-weight Access Request Form linked below. Access is reviewed annually in April.
These national systems, installed between 2016 and 2018, feature cutting-edge technology, and are available, free of charge, to researchers from post-secondary institutions.
Located at the University of Victoria, Arbutus is an OpenStack cloud, hosting virtual machines and other cloud workloads. The system, provided by Lenovo, has 14,968 CPU cores across 456 nodes, 146,944 GBs of RAM and accesses 5.7 PBs of persistent Ceph storage. Note that other cloud resources are available for a combined total of 17,272 cores nationally.
Located at the École de technologie supérieure in Montreal, Béluga is a general purpose cluster with 34,880 cores. Béluga, composed of a variety of nodes – including large memory nodes and GPU nodes with Turbo Boost activated – is designed to accommodate a broad range of workloads.
Located at the University of Waterloo, Graham is a heterogeneous cluster, suitable for a variety of workloads. It has a small OpenStack partition, and includes local storage on nodes. Specifications include 41,548 CPU cores across a diverse set of node types, including GPU nodes. The Graham system is entirely liquid cooled, using rear-door heat exchangers.
Located at the University of Toronto, Niagara has 80,640 CPU cores. This system has 202 GBs (188 GiB) of RAM per node and offers 7 PBs of scratch and 2 PBs of project storage space. Mellanox EDR InfiniBand is used to create a Dragonfly+ network topology featuring adaptive routing to provide the high-speed low-latency communications necessary for large-scale full-system simulations. A 256 TB Burst-buffer in this cluster helps improve performance for data-intensive work loads.