User Tools

Site Tools


cluster_billing

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
cluster_billing [2016/12/14 16:31]
mgstauff [Billing Cycle]
cluster_billing [2017/01/05 19:18] (current)
mgstauff [Storage]
Line 16: Line 16:
 Disk space is organized into project trees, consisting of a directory and all its sub-directories (e.g. /data/jet/mgstauff). Each project tree is assigned to a single //Billing Entity//. A billing entity is responsible for all project trees assigned to it, regardless of file ownership within the tree. Disk space is organized into project trees, consisting of a directory and all its sub-directories (e.g. /data/jet/mgstauff). Each project tree is assigned to a single //Billing Entity//. A billing entity is responsible for all project trees assigned to it, regardless of file ownership within the tree.
  
-Project trees can be setup as a shared directory for all users within a lab and/or within a project. Or, project trees can be setup for each user within a lab or project. However, the shared tree approach is preferred for easier management, maintenance and to facillitate file sharing within labs and projects.+Project trees can be setup as a shared directory for all users within a lab and/or within a project. Or, project trees can be setup for each user within a lab or project. However, the shared tree approach is preferred for easier management, maintenance and to facilitate file sharing within a lab and project.
  
 All storage is on dedicated enterprise-grade hardware RAIDs utilizing RAID-6 for redundancy. All storage is on dedicated enterprise-grade hardware RAIDs utilizing RAID-6 for redundancy.
 +
 +You may want to archive your data off of the cluster to save costs. If so, we have [[dataarchving|some suggestions here]]
  
 === Backup === === Backup ===
  
-All storage is backed up to tape on a quarterly basis and tapes are stored off-site. Users are responsible for maintaining their own copies of original data in the event of catastrophic failure of the system.+All storage is backed up to tape on a near-quarterly basis (at least 3 times per year) and tapes are stored off-site. __Users are responsible for maintaining their own copies of original data in the event of catastrophic failure of the system.__
  
 === Non-cluster Archiving & Cold Storage === === Non-cluster Archiving & Cold Storage ===
Line 31: Line 33:
 One slot is a quota of 1 cpu core and 6GB RAM (3GB on the 'basic' compute nodes). One slot is a quota of 1 cpu core and 6GB RAM (3GB on the 'basic' compute nodes).
  
-Each //Project Slot Quota// sets an aggregate slot quota for all users assigned to the //Project Slot Quota//. This quota limits the aggregate number of concurrent slots used by all the users in the //Project Slot Quota//. There is an individual limit of 50 slots for each user with a //Project Slot Quota//. Different individual quotas can be assigned for each user if needed, e.g. so a power user can be guaranteed more slots than the casual users in a group.+Each //Project Slot Quota// sets an aggregate slot quota for all users assigned to the //Project Slot Quota//. This quota limits the aggregate number of concurrent slots used by all the users in the //Project Slot Quota//Currently there is a maximum of 200 slots per __Billing Entity__. 
 + 
 +There is an individual limit of 40 slots for each user with a //Project Slot Quota//. Different individual quotas can be assigned for each user if needed, e.g. so a power user can be guaranteed more slots than the casual users in a group.
  
 For example: For example:
Line 39: Line 43:
 ^Aggregate quota|80| ^Aggregate quota|80|
 ^mgstauff quota|20| ^mgstauff quota|20|
-^pcook, jtduda quota|50|+^pcook, jtduda quota|40|
  
-In this example, the three users can together never use more than 80 slots at the same time. In addition, user mgstauff can never use more than 20, and users pcook and jtduda can each never user more than 50. Note that even though pcook and jtduda have 50-slot quotas, if pcook were already using 50 slots, then jtduda could only use 30 slots with any jobs he submitted at that time.+In this example, the three users can together never use more than 80 slots at the same time. In addition, user mgstauff can never use more than 20, and users pcook and jtduda can each never user more than 40. Note that even though pcook and jtduda have 40-slot quotas, if pcook were already using 40 slots, and mgstauff were using 15, then jtduda could only use 25 slots with any jobs he submitted at that time.
  
 A //Billing Entity// may decide to create individual //Project Slot Quotas// for power users, so they are not limited by an overall group slot quota. A //Billing Entity// may decide to create individual //Project Slot Quotas// for power users, so they are not limited by an overall group slot quota.
Line 67: Line 71:
 It provides: It provides:
   * 150 GB Disk Space   * 150 GB Disk Space
-  * One //Project Slot Quota// with a 16-core quota on the //Basic Cluster Queue//. This queue consists of about 240 cpu cores (slots) on older compute nodes (from the PICSL cluster) with less memory (16GB / machine) and a slower connection to the cluster frontend and data storage (1Gbit vs 10Gbit).+  * One //Basic Project Slot Quota// with a 16-core quota on the //Basic Cluster Queue//. This queue consists of about 240 cpu cores (slots) on older compute nodes (from the PICSL cluster) with less memory (16GB / machine) and a slower connection to the cluster frontend and data storage (1Gbit vs 10Gbit).
   * Unlimited number of assigned users   * Unlimited number of assigned users
   * Additional disk space is priced as described below.   * Additional disk space is priced as described below.
Line 73: Line 77:
 It does not provide: It does not provide:
   * Flexible per-user sub-quotas   * Flexible per-user sub-quotas
-  * Dyanmic quotas+  * Dynamic quotas
  
 === Storage - $700 / TB / year ( $400 / TB / year above 8TB ) === === Storage - $700 / TB / year ( $400 / TB / year above 8TB ) ===
Line 81: Line 85:
   * Multiple data directories can be assigned to a single //Billing Entity//   * Multiple data directories can be assigned to a single //Billing Entity//
   * Available to both Basic Accounts and to //Billing Entities// that opt for High-Speed Project Slot Quotas.   * Available to both Basic Accounts and to //Billing Entities// that opt for High-Speed Project Slot Quotas.
 +  * You may want to archive your data off of the cluster to save costs. If so, we have [[dataarchving|some suggestions here]]
  
 === High-speed Project Slot Quotas - $30 / slot / year === === High-speed Project Slot Quotas - $30 / slot / year ===
  
-  * Additional //Project Slot Quotas// give access to the //High-speed Cluster Queue// which utilizes newer high-speed, high-memory compute nodes (64GB), with fast (10Gbit) connection to storage. There is a total of 336 slots in this queue (cpu cores).+  * Additional //High-Speed Project Slot Quotas// give access to the //High-speed Cluster Queue// which utilizes newer high-speed, high-memory compute nodes (64GB), with fast (10Gbit) connection to storage. There is a total of 336 slots in this queue (cpu cores).
   * High-speed //Project Slot Quotas// are sized in 16-slot increments.   * High-speed //Project Slot Quotas// are sized in 16-slot increments.
   * Unlimited number of assigned users   * Unlimited number of assigned users
Line 91: Line 96:
   * Dynamic quotas -  every user's slot quota is increased when the cluster has idle computing power. Dynamic quota increases are proportional to each user's base quota.   * Dynamic quotas -  every user's slot quota is increased when the cluster has idle computing power. Dynamic quota increases are proportional to each user's base quota.
   * Quarterly billing for Project Slot Quotas is calculated using the largest number of slots allotted during the quarter.   * Quarterly billing for Project Slot Quotas is calculated using the largest number of slots allotted during the quarter.
 +  * [[slot_usage_reports|Slot Usage Reports]] are provided to help decide how many slots to request for a group.
  
   * NOTE: We have tried to design the Project Slot Quota system to be simple, flexible, affordable and to help researchers avoid the scenario in which they ask "Should I run this analysis again? How much is it going to cost me!?". We expect most labs and users to rarely change their Project Slot Quota assignments and quotas, perhaps once or twice a year at most. If we end up receiving many requests from a user to increase and decrease Project Slot Quota assignments, we will discuss the circumstances with the user to understand why it is necessary. That is, please don't try to game the system.   * NOTE: We have tried to design the Project Slot Quota system to be simple, flexible, affordable and to help researchers avoid the scenario in which they ask "Should I run this analysis again? How much is it going to cost me!?". We expect most labs and users to rarely change their Project Slot Quota assignments and quotas, perhaps once or twice a year at most. If we end up receiving many requests from a user to increase and decrease Project Slot Quota assignments, we will discuss the circumstances with the user to understand why it is necessary. That is, please don't try to game the system.
cluster_billing.1481733062.txt.gz · Last modified: 2016/12/14 16:31 by mgstauff