About Virtual Machines:
Academic Technology offers virtual machines (VMs) to support campus researchers, faculty, and students in their academic research and instructional activities.
We provide Linux VMs, with resource allocation that scales based on workload demands and the number of users. These VMs can be customized with varying amounts of processors, memory, and storage to meet the specific needs of different software applications and databases, optimizing system efficiency and performance. These VMs are ideal for tasks requiring moderate computational power.
Additionally, we offer a limited number of VMs equipped with vGPU resources, segmented from Nvidia A100s GPUs, available in memory slices ranging from 5GB to 40GB. These vGPUs are suitable for smaller workloads that do not need a full GPU. Due to the high demand, vGPUs are leased on a one-year lease term.
Our VMs are hosted on-campus datacenter with proper environmental controls and backup systems. Users can securely access these VMs via SSH, whether on-campus or remotely. However, a VPN connection is required if connecting remotely.
Requirements:
- VMs must be used for funded/sponsored research or for curriculum-based activities.
- Cannot be used to host Web content (i.e. websites or web applications).
- Faculty requesting VMs with GPU resources attached to them must agree with the 1-year lease term. Once the borrowed term ends the vGPU will be allocated to a different faculty member.
Recommended uses:
- Data analysis: processing datasets that are moderate in size using data analysis software (e.g. Python, R, MySQL).
- Scientific simulations: performing simulations to model real-world systems and processes.
- Machine learning: training and executing machine learning models on moderate-sized datasets.
If you are interested in acquiring a virtual machine for your research you may submit a request via ServiceNOW or by email at@sfsu.edu. This will usually lead to a brief introductory meeting with our systems team to discuss your research needs.
Comments
0 comments
Article is closed for comments.