38,760
edits
Technical glossary for the resource allocation competitions/en: Difference between revisions
Jump to navigation
Jump to search
Technical glossary for the resource allocation competitions/en (view source)
Revision as of 17:47, 30 August 2023
, 1 year agoUpdating to match new version of source page
(Updating to match new version of source page) |
(Updating to match new version of source page) |
||
| Line 8: | Line 8: | ||
'''VGPU:''' Stands for virtual graphics processing unit (VGPU). One or more VGPUs can be assigned to Virtual Machines (VM) within a cloud environment. Each VGPU is seen as a single physical GPU device by the VM's operating system. | '''VGPU:''' Stands for virtual graphics processing unit (VGPU). One or more VGPUs can be assigned to Virtual Machines (VM) within a cloud environment. Each VGPU is seen as a single physical GPU device by the VM's operating system. | ||
'''Reference GPU Unit (RGU):''' RGU is a unit measuring the amount of GPU resources that are used. It represents the "cost" of utilizing a particular GPU model, whose RGU value varies based on performance. For example: 1 GPU A100-40GB = 4.0 RGU; 1 GPU V100-16GB = 2.2 RGU; 1 GPU P100-12GB = 1.0 RGU. | |||
== Resource allocations == | == Resource allocations == | ||
| Line 23: | Line 25: | ||
'''Core equivalent:''' A core equivalent is a bundle made up of a single core and some amount of associated memory. In other words, a core equivalent is a core plus the amount of memory considered to be associated with each core on a given system. See detailed explanation [[Allocations_and_compute_scheduling|here]]. | '''Core equivalent:''' A core equivalent is a bundle made up of a single core and some amount of associated memory. In other words, a core equivalent is a core plus the amount of memory considered to be associated with each core on a given system. See detailed explanation [[Allocations_and_compute_scheduling|here]]. | ||
'''GPU year:''' a GPU year is the equivalent of using 1 GPU continuously for a full year or 12 GPU for a month. | |||
'''RGU year:''' RGU year is a calculated value that results from multiplying GPU years times the RGU of a given GPU model. For example, 10 GPU years of an A100-40GB (which ''costs'' 4 RGU) equals 40 RGU years. | |||
'''Head or Login node:''' Typically when you access a cluster you are accessing a head node, or gateway/login node. A head node is configured to be the launching point for jobs running on the cluster. When you are told or asked to login or access a cluster, invariably you are being directed to log into the head node, often nothing more than a node configured to act as a middle point between the actual cluster and the outside network. | '''Head or Login node:''' Typically when you access a cluster you are accessing a head node, or gateway/login node. A head node is configured to be the launching point for jobs running on the cluster. When you are told or asked to login or access a cluster, invariably you are being directed to log into the head node, often nothing more than a node configured to act as a middle point between the actual cluster and the outside network. | ||
| Line 41: | Line 47: | ||
'''Memory per node:''' The total amount of installed RAM in a compute node. | '''Memory per node:''' The total amount of installed RAM in a compute node. | ||
''' | '''System memory per core:''' Amount of system memory requested by a job in GB divided by the number of CPU cores requested by the job. | ||
''' | '''System memory per GPU:''' Amount of system memory requested by a job in GB divided by the number of GPUs requested by the job. | ||
== Storage == | == Storage == | ||
| Line 68: | Line 74: | ||
== Cloud == | == Cloud == | ||
'''Alliance Cloud:''' is a pool of hardware supporting virtualization. This can be thought of as Infrastructure as a Service (IaaS). | '''Alliance Cloud:''' is a pool of hardware supporting virtualization. This can be thought of as Infrastructure as a Service (IaaS). | ||
'''VCPU year:''' same as CPU year, but for cloud. | |||
'''VGPU year:''' same as GPU year, but for cloud. | |||
'''Compute instances:''' These are instances that have a limited life-time and typically have constant high-CPU requirements for the instances life-time. They have also been referred to as ‘batch’ instances. These will be granted higher vCPU/Memory quotas since they are time-limited instances. | '''Compute instances:''' These are instances that have a limited life-time and typically have constant high-CPU requirements for the instances life-time. They have also been referred to as ‘batch’ instances. These will be granted higher vCPU/Memory quotas since they are time-limited instances. | ||
'''Persistent instances:''' These are instances that are meant to run indefinitely (e.g., based on the clouds availability) and would include web servers, database servers, etc. In general, these are thought to be lower CPU or bursty CPU instances. These will have lower vCPU/Memory quotas since they are meant to consume the resources for long periods of time. | '''Persistent instances:''' These are instances that are meant to run indefinitely (e.g., based on the clouds availability) and would include web servers, database servers, etc. In general, these are thought to be lower CPU or bursty CPU instances. These will have lower vCPU/Memory quotas since they are meant to consume the resources for long periods of time.<br> | ||
'''Cloud storage:''' Persistent cloud storage provides virtual disk functionality to virtual machines running in the cloud. Persistent cloud storage is very reliable and scalable, made possible by specialized software (Ceph). | '''Cloud storage:''' Persistent cloud storage provides virtual disk functionality to virtual machines running in the cloud. Persistent cloud storage is very reliable and scalable, made possible by specialized software (Ceph). | ||