Anvil from Purdue University is an NSF Category I (Award #2005632) computational resource with a comprehensive ecosystem of software, access interfaces, programming environments, and composable services in a seamless environment to support a broad range of current and future science and engineering applications. When deployed, Anvil will consist of 1000 128-core CPU nodes based on AMD's Milan architecture, 32 large memory nodes with 1 TB of RAM each, and 16 nodes of Nvidia Volta Next GPUs (16x4 total GPUs, 40GB memory/GPU), along with a multitier storage system including long-term archival, persistent file and campaign storage, a scratch file system, a burst buffer, and object storage to support a variety of workflows and storage needs.
Anvil will be connected to the XSEDEnet via a 200 Gbps network link to the Indiana GigaPOP and will support Globus for high-speed, reliable, and secure data transfer. Anvil will also feature a composable subsystem supporting Cloud and container-based workflows and integrated pathways to Microsoft Azure helping researchers leverage both on-premises and commercial Cloud computing for advanced applications. Anvil will provide approximately one billion CPU core hours per year to XSEDE users for five years. Anvil GPUs, large memory nodes, and the services of its composable subsystems will also be provided to XSEDE users through the allocation process. The Anvil system is tentatively scheduled to begin an early user phase in late summer of 2021 and enter production by October 1, 2021.
Johns Hopkins University, through the Maryland Advanced Research Computing Center (MARCC), will participate in the XSEDE Federation with its new NSF-funded flagship cluster "rockfish.jhu.edu" funded by NSF MRI award #1920103 that integrates high-performance and data-intensive computing while developing tools for generating, analyzing and disseminating data sets of ever-increasing size. The cluster will contain compute nodes optimized for different research projects and complex, optimized workflows. Rockfish consists of 368 regular compute nodes with 192GB of memory, 10 large memory nodes with 1,5TB of memory and 10 GPU nodes with 4 Nvidia A100 GPUs featuring Intel Cascade Lake 6248R, 48 cores per node, 3,0GHz processor base frequency, and 1TB NVMe for local storage. All compute nodes have HDR100 connectivity. In addition, the cluster has access to several GPFS file systems totaling 10PB of storage. 20% of these resources will be allocated via XSEDE. This system is tentatively scheduled to join the XSEDE community in summer 2021.
Texas A&M will be offering FASTER - Fostering Accelerated Scientific Transformations, Education and Research, a composable high-performance data-analysis and computing instrument. FASTER was funded by the NSF MRI programme (Award #2019129). The FASTER system will have about 180 compute nodes with 2 Intel 32-core Ice Lake processors and include about 270 NVIDIA GPUs (40 A100 and 230 T4/A40/A10 GPUs). Using LIQID's composable technology, all 180 compute nodes will have access to the pool of available GPUs, dramatically improving workflow scalability. FASTER will have HDR InfiniBand interconnection and access/share a 5PB usable high-performance storage system running Lustre filesystem. 30% of FASTER's computing resources will be allocated to researchers nationwide through XSEDE's XRAC process. This system is tentatively scheduled to go into production in early summer 2021.
The University of Kentucky will provide XSEDE users with access to the large-memory computational resource KyRIC - Kentucky Research Informatics Cloud. KyRIC was funded by the NSF MRI programme (Award #1626364) and will advance several exciting research programs across many disciplines, such as bioinformatics and system biology algorithms, large graph and evolutionary network analysis, image processing, and computational modelling and simulation. The KyRIC system is a hybrid architecture. KyRIC large memory nodes allocable by XSEDE will provide 3TB of shared memory for processing massive NLP data sets, genome sequencing, bioinformatics and memory intensive analysis of Big Data. Each of KyRIC's 5 large memory nodes will consist of Intel Xeon CPU E7-4820 v4 @ 2.00GHz with 4 sockets, 10 cores/socket, 3TB RAM, 6TB SSD storage drives and 100G Ethernet interconnects. This system is currently available to be requested in the current allocation submission cycle for allocations starting in April 2021.
XSEDE also welcomes the Open Storage Network (OSN), a distributed data sharing and transfer service intended to facilitate exchanges of active scientific data sets between research organizations, communities and projects, providing easy access and high bandwidth delivery of large data sets to researchers who leverage the data to train machine learning models, validate simulations, and perform statistical analysis of live data. OSN is funded by NSF Collaboration Awards #1747507, #1747490, #1747483, #1747552, and #1747493. The OSN is intended to serve two principal purposes: (1) enable the smooth flow of large data sets between resources such as instruments, campus data centres, national supercomputing centres, and Cloud providers; and (2) facilitate access to long tail data sets by the scientific community. Examples of data currently available on the OSN include synthetic data from ocean models; the widely used Extracted Features Set from the Hathi Trust Digital Library; open access earth sciences data from Pangeo; and Geophysical Data from BCO-DMO.
OSN data is housed in storage pods, located at Big Data Hubs, interconnected by national, high-performance networks and accessed via a RESTful interface following Simple Storage System (S3) conventions, creating well-connected, Cloud-like storage with data transfer rates comparable to or exceeding the public cloud storage providers, where users can park data, back data up, and/or create readily accessible storage for active data sets. 5 PB of storage are currently available for allocation. Allocations of a minimum 10 TB and max of 300 TB can be requested through the XRAC process. XSEDE researchers can ask for OSN allocations in this current allocation request cycle that ends January 15.